Description:FORM 2

THE PATENTS ACT, 1970
(39 of 1970)
&
THE PATENT RULES, 2003

COMPLETE SPECIFICATION
(See Section 10 and Rule 13)

Title of invention:
MACHINE LEARNING (ML) BASED RECOMMENDATION FOR DATA CENTER MIGRATION

Applicant
Tata Consultancy Services Limited
A company Incorporated in India under the Companies Act, 1956
Having address:
Nirmal Building, 9th floor,
Nariman point, Mumbai 400021,
Maharashtra, India

Preamble to the description:

The following specification particularly describes the invention and the manner in which it is to be performed.
TECHNICAL FIELD
[001] The embodiments herein generally relate to the field of data migration and, more particularly, to a method and system for Machine Learning (ML) based recommendation for data center migration.

BACKGROUND
[002] Data center migration is the process of moving select assets from one data center environment to another. Data center migration brings many advantages as it highlights any redundancies in data, allows consolidation of data and processes, and removes unnecessary data, means more space is available on server(s). Depending on the situation, this can mean a reduction in servers maintained, space rented, or even physical locations of data centers.
[003] However, all recent developments in automation of data migration focus totally on a cloud migration technology that assists organizations in migrating their data, applications, and infrastructure from on-premises to the cloud. Azure, S/4HANA, AWS are common providers of cloud platforms for data migration, and they all utilize Machine Learning (ML) capabilities in one or the other way. However, mode of deployment/operation (web-based/ cloud-based) can be a differentiator.
[004] Existing cloud migration approaches are not directly suitable for data center migration for reasons such as poor knowledge of processing data which leads to critical failures. There are many existing tools for data migration, but to make effective use of them, data center migration tasks requires certain approaches to be followed. Datacenters migration has many complexities such as disruptions by unexpected challenges, lack of pre-migration planning, lack of envision of post migration environment, lack of skill to perform migration, improper backup schedule etc. Current approaches are entirely based on manual intervention. An automation that can handle the above complexities of data center migration needs to be explored.

SUMMARY
[005] Embodiments of the present disclosure present technological improvements as solutions to one or more of the above-mentioned technical problems recognized by the inventors in conventional systems.
[006] For example, in one embodiment, a method for Machine Learning (ML) based recommendation for data center migration is provided. The method includes performing data discovery from a master Configuration Item (CI) list, a master discovery list, event updates and bulk migration updates, wherein a discovered data comprises a plurality of parameters associated with an existing infrastructure of a datacenter. Further, the method includes pre-processing the plurality of parameters. Further, the method includes segregating the pre-processed plurality of parameters into a plurality of groups, wherein each group among the plurality of groups is associated with a pretrained Machine Learning (ML) model among a plurality of pretrained ML models based on a predefined relevancy of each of the preprocessed plurality of parameters with each of the plurality of pretrained ML models the plurality of pretrained ML models comprising a treatment type recommender, a T-shirt sizing recommender, a big rule recommender, a move group recommender, And an event planner-scheduler. Further, the method includes extracting a set of optimum features, for each of the plurality of pretrained ML models, from the segregated plurality of parameters of the group based on a reference library and domain knowledge. Furthermore, the method includes generating, for a migration event of the datacenter towards a target environment, a set of recommendations by each of the plurality of pretrained ML models based on the set of features extracted for each of the plurality of pretrained model. The treatment type recommender recommends a pair of treatment types and associated pattern indicating a migration approach to be used, the T-shirt sizing recommender bundles a fixed amount of RAM and storage with a given number of virtual CPU cores based on need for scale-up or scale-down of compute resources post migration, the big rule recommender classifies a plurality of migration approaches into one of a BigRule violation and a BigRule exceptions, providing checkpoint to finalize the migration approach recommended by the treatment type recommender that enables to plan and schedule the migration accordingly, the move group recommender recommends a move group among a plurality of move groups, wherein the move group decides the migration event based applications, subnet or environment types that can migrate together to a target environment, and the event planner-scheduler recommends i) the planning and scheduling of one or more events of the migration event, scale out and scale down based on available bandwidth of resources, based on replication appliance available, ii) for a recommended number of servers further recommends resources, duration, weekly schedule, and monthly schedule, and iii) optimization of upcoming one or more events based on event analysis to increase velocity and success rate of the migration event. Further, the method includes triggering a migration tool for the migration event to be initiated for migration of the datacenter to the target environment in accordance with the set of recommendations of each of the plurality of pretrained ML models.
[007] In another aspect, a system for Machine Learning (ML) based recommendation for data center migration is provided. The system comprises a memory storing instructions; one or more Input/Output (I/O) interfaces; and one or more hardware processors coupled to the memory via the one or more I/O interfaces, wherein the one or more hardware processors are configured by the instructions to
[008] In yet another aspect, there are provided one or more non-transitory machine-readable information storage mediums comprising one or more instructions, which when executed by one or more hardware processors causes a method for Machine Learning (ML) based recommendation for data center migration.
[009] The one or more hardware processors causes performing data discovery from a master Configuration Item (CI) list, a master discovery list, event updates and bulk migration updates, wherein a discovered data comprises a plurality of parameters associated with an existing infrastructure of a datacenter. Further, the method includes pre-processing the plurality of parameters. Further, the one or more hardware processors causes segregating the pre-processed plurality of parameters into a plurality of groups, wherein each group among the plurality of groups is associated with a pretrained Machine Learning (ML) model among a plurality of pretrained ML models based on a predefined relevancy of each of the preprocessed plurality of parameters with each of the plurality of pretrained ML models the plurality of pretrained ML models comprising a treatment type recommender, a T-shirt sizing recommender, a big rule recommender, a move group recommender, And an event planner-scheduler. Further, the method includes extracting a set of optimum features, for each of the plurality of pretrained ML models, from the segregated plurality of parameters of the group based on a reference library and domain knowledge. Furthermore, the one or more hardware processors causes generating, for a migration event of the datacenter towards a target environment, a set of recommendations by each of the plurality of pretrained ML models based on the set of features extracted for each of the plurality of pretrained model. The treatment type recommender recommends a pair of treatment types and associated pattern indicating a migration approach to be used, the T-shirt sizing recommender bundles a fixed amount of RAM and storage with a given number of virtual CPU cores based on need for scale-up or scale-down of compute resources post migration, the big rule recommender classifies a plurality of migration approaches into one of a BigRule violation and a BigRule exceptions, providing checkpoint to finalize the migration approach recommended by the treatment type recommender that enables to plan and schedule the migration accordingly, the move group recommender recommends a move group among a plurality of move groups, wherein the move group decides the migration event based applications, subnet or environment types that can migrate together to a target environment, and the event planner-scheduler recommends i) the planning and scheduling of one or more events of the migration event, scale out and scale down based on available bandwidth of resources, based on replication appliance available, ii) for a recommended number of servers further recommends resources, duration, weekly schedule, and monthly schedule, and iii) optimization of upcoming one or more events based on event analysis to increase velocity and success rate of the migration event. Further, the one or more hardware processors causes triggering a migration tool for the migration event to be initiated for migration of the datacenter to the target environment in accordance with the set of recommendations of each of the plurality of pretrained ML models.
[0010] The method includes performing data discovery from a master Configuration Item (CI) list, a master discovery list, event updates and bulk migration updates, wherein a discovered data comprises a plurality of parameters associated with an existing infrastructure of a datacenter. Further, the method includes pre-processing the plurality of parameters. Further, the method includes segregating the pre-processed plurality of parameters into a plurality of groups, wherein each group among the plurality of groups is associated with a pretrained Machine Learning (ML) model among a plurality of pretrained ML models based on a predefined relevancy of each of the preprocessed plurality of parameters with each of the plurality of pretrained ML models the plurality of pretrained ML models comprising a treatment type recommender, a T-shirt sizing recommender, a big rule recommender, a move group recommender, And an event planner-scheduler. Further, the method includes extracting a set of optimum features, for each of the plurality of pretrained ML models, from the segregated plurality of parameters of the group based on a reference library and domain knowledge. Furthermore, the method includes generating, for a migration event of the datacenter towards a target environment, a set of recommendations by each of the plurality of pretrained ML models based on the set of features extracted for each of the plurality of pretrained model. The treatment type recommender recommends a pair of treatment types and associated pattern indicating a migration approach to be used, the T-shirt sizing recommender bundles a fixed amount of RAM and storage with a given number of virtual CPU cores based on need for scale-up or scale-down of compute resources post migration, the big rule recommender classifies a plurality of migration approaches into one of a BigRule violation and a BigRule exceptions, providing checkpoint to finalize the migration approach recommended by the treatment type recommender that enables to plan and schedule the migration accordingly, the move group recommender recommends a move group among a plurality of move groups, wherein the move group decides the migration event based applications, subnet or environment types that can migrate together to a target environment, and the event planner-scheduler recommends i) the planning and scheduling of one or more events of the migration event, scale out and scale down based on available bandwidth of resources, based on replication appliance available, ii) for a recommended number of servers further recommends resources, duration, weekly schedule, and monthly schedule, and iii) optimization of upcoming one or more events based on event analysis to increase velocity and success rate of the migration event. Further, the method includes triggering a migration tool for the migration event to be initiated for migration of the datacenter to the target environment in accordance with the set of recommendations of each of the plurality of pretrained ML models.
[0011] It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention, as claimed.

BRIEF DESCRIPTION OF THE DRAWINGS
[0012] The accompanying drawings, which are incorporated in and constitute a part of this disclosure, illustrate exemplary embodiments and, together with the description, serve to explain the disclosed principles:
[0013] FIG. 1A is a functional block diagram of a system, for Machine Learning (ML) based recommendation for data center migration, in accordance with some embodiments of the present disclosure.
[0014] FIG. 1B illustrates an architectural overview of the system of FIG. 1A, in accordance with some embodiments of the present disclosure.
[0015] FIGS. 2A through 2B (collectively referred as FIG. 2) is a flow diagram illustrating a method for Machine Learning (ML) based recommendation for data center migration, using the system depicted in FIG. 1A and 1B, in accordance with some embodiments of the present disclosure.
[0016] FIG. 3 depicts a detailed process workflow of the system, in accordance with some embodiments of the present disclosure.
[0017] FIG. 4 depicts a Deep Learning (DL) model for treatment type recommendations, in accordance with some embodiments of the present disclosure.
[0018] FIG. 5 depicts a Decision tree and a random forest combined using ensemble method for training t-shirt sizing and defining move group recommendations, in accordance with some embodiments of the present disclosure.
[0019] FIG. 6 depicts discovery data in tabular format.
[0020] It should be appreciated by those skilled in the art that any block diagrams herein represent conceptual views of illustrative systems and devices embodying the principles of the present subject matter. Similarly, it will be appreciated that any flow charts, flow diagrams, and the like represent various processes which may be substantially represented in computer readable medium and so executed by a computer or processor, whether or not such computer or processor is explicitly shown.

DETAILED DESCRIPTION OF EMBODIMENTS
[0021] Exemplary embodiments are described with reference to the accompanying drawings. In the figures, the left-most digit(s) of a reference number identifies the figure in which the reference number first appears. Wherever convenient, the same reference numbers are used throughout the drawings to refer to the same or like parts. While examples and features of disclosed principles are described herein, modifications, adaptations, and other implementations are possible without departing from the scope of the disclosed embodiments.
[0022] Existing cloud migration approaches are not directly suitable for data center migration due to the complexities involved with data centers as explained in the background section. There are many existing tools for data migration, but to make effective use of them, data center migration tasks requires certain approaches to be followed. Current approaches are entirely based on manual intervention, whereas automation in this domain of data center migration needs to be explored. Workload migration projects tend to exceed budget as well as time duration and result in business disruption due to lack of data accuracy, lack of visibility and transparency on overall discovery migration and decommission phases across distributed teams.
[0023] There are problems of failures in Infrastructure discovery and migration due to lack of knowledge in processing various data sources, which includes duplicate data, missing data, erroneous mapping, erroneous planning, erroneous sizing leads to inappropriate data processing or incomplete migration planning to generate the trusted data as input for migration planning. There is lack of customized tools to integrate the 3rd party discovery tools and integrated process to take decisions relevant to migration project which leads to delays in overall migration project. There are no direct tools to collaborate at granular level including move group creation, runbook activity, tasks list, workflow, status tracking. Lots of time and effort is consumed to generate the migration status reports, and dashboard. There is no solution to easily predict the timeline, velocity and resources required to execute the migration based on the history of previous events using spreadsheets or mail or documents.
[0024] Embodiments of the present disclosure provide a method and system for Machine Learning (ML) based recommendation for data center migration. The system, interchangeably referred to as an infrastructure migration cockpit, generates ML based recommendations to address and reduce complexity around the datacenter migration, which accelerates the migration planning and improve efficiency of the end-to-end migration lifecycle. The system manages the discovery and migration seamlessly by addressing the following challenges.
[0025] The system disclosed provides a tool to manage the multiple migration events, automate manual efforts to generate various types of reports and dashboards to review and validate the status of migration events, generating move group creating suggestion based on technical and other inputs, suggesting for resource optimization based on big rule inputs, recommending for migration pattern, and improving the quality of discovery and migration execution.
[0026] Referring now to the drawings, and more particularly to FIGS. 1A through 6, where similar reference characters denote corresponding features consistently throughout the figures, there are shown preferred embodiments and these embodiments are described in the context of the following exemplary system and/or method.
[0027] FIG. 1A is a functional block diagram of a system, for Machine Learning (ML) based recommendation for data center migration, in accordance with some embodiments of the present disclosure.
[0028] In an embodiment, the system 100 includes a processor(s) 104, communication interface device(s), alternatively referred as input/output (I/O) interface(s) 106, and one or more data storage devices or a memory 102 operatively coupled to the processor(s) 104. The system 100 with one or more hardware processors is configured to execute functions of one or more functional blocks of the system 100.
[0029] Referring to the components of system 100, in an embodiment, the processor(s) 104, can be one or more hardware processors 104. In an embodiment, the one or more hardware processors 104 can be implemented as one or more microprocessors, microcomputers, microcontrollers, digital signal processors, central processing units, state machines, logic circuitries, and/or any devices that manipulate signals based on operational instructions. Among other capabilities, the one or more hardware processors 104 are configured to fetch and execute computer-readable instructions stored in the memory 102. In an embodiment, the system 100 can be implemented in a variety of computing systems including laptop computers, notebooks, hand-held devices such as mobile phones, workstations, mainframe computers, servers, and the like.
[0030] The I/O interface(s) 106 can include a variety of software and hardware interfaces, for example, a web interface, a graphical user interface and the like and can facilitate multiple communications within a wide variety of networks N/W and protocol types, including wired networks, for example, LAN, cable, etc., and wireless networks, such as WLAN, cellular and the like. In an embodiment, the I/O interface (s) 106 can include one or more ports for connecting to a number of external devices or to another server or devices.
[0031] The memory 102 may include any computer-readable medium known in the art including, for example, volatile memory, such as static random access memory (SRAM) and dynamic random access memory (DRAM), and/or non-volatile memory, such as read only memory (ROM), erasable programmable ROM, flash memories, hard disks, optical disks, and magnetic tapes.
[0032] In an embodiment, the memory 102 includes a plurality of modules 110. For example, as depicted in FIG. 1B, the modules can include a plurality of ML models comprising a treatment type recommender, a T-shirt sizing recommender, a big rule recommender, and a move group recommender. and an event planner-scheduler and so on. The training and the ML techniques used for each of the ML models is explained in conjunction with FIGS. 4 and 5.
[0033] The plurality of modules 110 include programs or coded instructions that supplement applications or functions performed by the system 100 for executing different steps involved in the process of Machine Learning (ML) based recommendation for data center migration, being performed by the system 100. The plurality of modules 110, amongst other things, can include routines, programs, objects, components, and data structures, which performs particular tasks or implement particular abstract data types. The plurality of modules 110 may also be used as, signal processor(s), node machine(s), logic circuitries, and/or any other device or component that manipulates signals based on operational instructions. Further, the plurality of modules 110 can be used by hardware, by computer-readable instructions executed by the one or more hardware processors 104, or by a combination thereof. The plurality of modules 110 can further include various sub-modules (not shown). Further, the memory 102 may comprise information pertaining to input(s)/output(s) of each step performed by the processor(s) 104 of the system100 and methods of the present disclosure.
[0034] Further, the memory 102 includes a database 108. The database (or repository) 108 may include a plurality of abstracted piece of code for refinement and data that is processed, received, or generated as a result of the execution of the plurality of modules in the module(s) 110, such as recommendations generated by the plurality of ML models.
[0035] Although the database 108 is shown internal to the system 100, it will be noted that, in alternate embodiments, the database 108 can also be implemented external to the system 100, and communicatively coupled to the system 100. The data contained within such an external database may be periodically updated. For example, new data may be added into the database (not shown in FIG. 1A) and/or existing data may be modified and/or non-useful data may be deleted from the database. In one example, the data may be stored in an external system, such as a Lightweight Directory Access Protocol (LDAP) directory and a Relational Database Management System (RDBMS). Functions of the components of the system 100 are now explained with reference to FIG. 1B through FIG. 5.
[0036] FIG. 1B illustrates an architectural overview of the system of FIG. 1A and is further explained in accordance with FIG. 2B through FIG. 5. The system 100 or the infrastructure migration cockpit brings an intelligent and user-friendly environment to manage the Infrastructure discovery data, governance the migration and decommission activities efficiently. This infrastructure migration cockpit is a web-based tool providing facility to manage the discovery data in a given data center environment to keep as the highest trusted source for migration planning activities such as validating the discovery data, gap analysis, move group creation based on affinity, create pre-migration, migration, post migration tasks in quality gates. the feature helps to quickly generate discovery reports and dashboards based on custom requirements, The tool is configured with various technical and non-technical inputs which can suggest the recommended migration move groups based on affinity and dependencies to be migrated together. The system 100 helps to identify the resource optimization in target based on the big rule inputs. Thus, provides governance for discovery, planning, and migration execution.
[0037] As depicted in FIG. 1B, the system 100 performs data discovery, which otherwise tremendous manual efforts for collecting the various data sources like Rvtools report, Configuration Management Database (CMDb) report, Inventory files and transport the highest trusted data into excel files to prepare the final scope validation, capacity planning. The system 100 collectively captures the required configuration details from various data sources which are critical for migration planning activities and creates a single master discovery sheet to populate the data and consume for migration planning with all the critical fields. The system 100 also has an API integration feature enabled to directly collect the discovery data. It can then generate the master inventory file which can be used as CMDB for customers. The system 100 performs data discovery using two approaches.
[0038] In the first approach, discover data is pushed from external discovery tools using Rest API like Device 42 etc.
[0039] In the second approach library module is generated to extract data using shell and PowerShell scripts. The library module has below features:
a) Captures config data from windows, Linux & Solaris, multiple servers at a time.
b) Stores credentials in local machine and encrypted in the script while connecting to target server modular approach with calling script and called scripts (12) and easy to maintain the code.
[0040] Discover data has ‘data’ as provided in Tabular format of FIG. 6.
[0041] The system 100 includes ML based recommendation modules, that are trained based on parameters extracted from discovery data. The system 100 utilizes known in the art bagging methods with random number to perform extraction of the parameters.
[0042] The treatment type recommender: The treatment type recommender recommends a pair of treatment types and associated pattern indicating a migration approach to be used, Thus, predicts the migration pattern (R-Factor Treatment type recommendation). Provides recommendations on migration patterns such as P2V, V2V, Lift and shift, File migration, Block migration, Database migration based on big rules. Platform recommendation model is key to choose the correct migration pattern, this helps to select migration tools, move group plan, migration execution in advance.
[0043] The move group recommender: Identifying the applications and infrastructure that can move to the target environment is called move group, this move group consists of multiple workloads, which are part of the same application or subnet or environment and grouped together according to the defined timelines. These move groups are identical to both production and non-production environments. Move group recommendation aims to automate the manual task of recommending the move group using ML.
[0044] Event planner and scheduler: Handles governance on migration execution. The pre-migration, migration cutover and post migration activities can be created and assigned to any individuals for multiple events. This helps to achieve the following.
1.Centralized Dashboard Simplifies Collaboration & Management
2.Workflow based assignment of task upon completion of previous task
3.Granular tracking across stages of transformation & near real time report extraction of ongoing project status.
4.Eliminating efforts of creating multiple trackers to keep track of progress
[0045] FIGS. 2A through 2B (collectively referred as FIG. 2) is a flow diagram illustrating a method 200 for Machine Learning (ML) based recommendation for data center migration, using the system depicted in FIG. 1A and 1B, in accordance with some embodiments of the present disclosure.
[0046] In an embodiment, the system 100 comprises one or more data storage devices or the memory 102 operatively coupled to the processor(s) 104 and is configured to store instructions for execution of steps of the method 200 by the processor(s) or one or more hardware processors 104. The steps of the method 200 of the present disclosure will now be explained with reference to the components or blocks of the system 100 as depicted in FIG. 1A and 1B and the steps of flow diagram as depicted in FIG. 2. Although process steps, method steps, techniques or the like may be described in a sequential order, such processes, methods, and techniques may be configured to work in alternate orders. In other words, any sequence or order of steps that may be described does not necessarily indicate a requirement that the steps be performed in that order. The steps of processes described herein may be performed in any order practical. Further, some steps may be performed simultaneously.
[0047] Referring to the steps of the method 200, at step 202 of the method 200, the one or more hardware processors 104 are configured to performing data discovery from a master Configuration Item (CI) list, a master discovery list, event updates and bulk migration updates. The discovered data comprises a plurality of parameters associated with an existing infrastructure of a datacenter.
[0048] At step 204 of the method 200, the one or more hardware processors 104 are configured to preprocess the plurality of parameters. This improves the accuracy and stability of statistical models and ML models, and in turn reduces the impact on results. It helps to get extreme values that differ from the rest of the data... The preprocessing includes following steps:
a) Discretization (grouping and sorting), - Converting attributes values of continuous data into a finite set of intervals with minimum data loss
b) Imputation (handling missing values) - substituting missing values to retain data and improve the training process
c) Feature splitting - Split is the process of splitting features intimately into two or more parts and performing to make new features. This technique helps the algorithms to better understand and learn the patterns in the dataset.
d) Handling outliers- outliers are the data points that can skew the average/mean and can negatively affect the statistical analysis resulting in lower accuracies
e) Categorical encoding (one hot encoding) - Transforms the categorical variable (parameter) into a set of binary variables
f) Variable transformation techniques that help with normalizing skewed data,
g) Scaling - Data scaling in python is an essential process to follow before modeling. The data within a similar scale can surprisingly increase the model’s predictive power. One of the most common scaling techniques is normalization where data values lie within a common scale (0 –1)
h) Correlation matrices. - It allows us to visualize how much (or how little) correlation exists between different variables.
[0049] At step 206 of the method 200, the one or more hardware processors 104 are configured to segregate the preprocessed plurality of parameters into a plurality of groups. Each group among the plurality of groups is associated with a pretrained Machine Learning (ML) model among the plurality of pretrained ML models (ML models). The plurality of parameters that are identified as significant features for each ML model is based on a predefined relevancy of each of the preprocessed plurality of parameters with each of the plurality of pretrained ML models. The ML models as depicted in FIG. 1B include the treatment type recommender, the T-shirt sizing recommender, the big rule recommender, and the move group recommender. and the event planner-scheduler. Each of the ML models are trained using acquired training dataset by extracting features extracted from the relevant plurality features belonging to associated group.
[0050] At step 208 of the method 200, the one or more hardware processors 104 are configured to extract a set of optimum features, for each of the plurality of pretrained ML models, from the segregated plurality of parameters of the group based on a reference library and domain knowledge.
[0051] At step 210 of the method 200, the one or more hardware processors 104 are configured to generate, for a migration event of the datacenter towards a target environment, a set of recommendations by each of the plurality of pretrained ML models based on the set of features extracted for each of the plurality of pretrained models.
[0052] The treatment type recommender: Recommends a pair of treatment types and associated pattern indicating a migration approach to be used. During training and inferencing of the plurality of pretrained ML model, the segregated plurality of parameters relevant to the treatment type recommender comprise server_Type, Server_Role, Operating_System, CPU_Total_Cores, Avg_CPU_Utilization_3months, Memory_RAM_GB, Avg_Memory_Utilization_3Months Total_Allocated_Capacity_GB, Total_Used_Capacity_GB, LUN, Storage_Name Server_ISCSI, Server_WWPN, No_of_NICs, Part_of_Cluster Shared_LUN, Shared_Cluster, NAS_Storage, and Mount_Location.
[0053] The above data is gathered and converted into Pandas data frame (data format for building a machine-learning model) using python script and Pandas library i.e., building the R3 Recommendation model or Treatment type recommendation model. The above-prepared data is then fed to a Decision tree model for generalization, estimations are calculated, and predictions will be generated with probability scores for each of the predictions. All the inferences generated for Treatment type recommendations, as provided in example Table 1 below, are updated directly on a Master Discovery update section located in the database 108 of the system 100.
Table 1:
S. No Treatment Types Pattern
1 Physical to Physical Re-host
2 Physical to Virtual Re-host
3 Virtual to Virtual Re-host
4 Lift and Shift Lift and Shift
5 Data Migration Build & Restore
6 SAN Migration Re-host/ Lift and shift
7 NAS Migration Build & Restore

[0054] The T-shirt sizing recommender: Bundles a fixed amount of RAM and storage with a given number of virtual CPU cores based on need for scale-up or scale-down of compute resources post migration. If more resource is needed say, RAM, for example –a larger sized workload instance must be consumed. These T-shirt sizing offerings are typically classified as small, medium, large. The system 100 aims to automate the “T-shirt Size Recommendation” process using ML models to confirm the CPU and memory sizes on target environment.
[0055] During training and inferencing of the plurality of pretrained ML model, the segregated plurality of parameters relevant to the T-shirt sizing recommender comprise Server_Type Server_Role Operating_System CPU_Total_Cores Avg_CPU_Utilization_3months, Memory_RAM_GB Avg_Memory_Utilization_3Months, Total_Allocated_Capacity_GB Total_Used_Capacity_GB LUN Storage_Name Server_ISCSI Server_WWPN No_of_NICs Part_of_Cluster Shared_LUN Shared_Cluster NAS_Storage Mount_Location;
[0056] Output of the model is provided below in Table 2.
Table 2:
Server Sizes Processor Memory (GB)
Small Servers
S1 1 2
S2 2 4
S3 2 8
S4 4 16
Medium Servers
M1 4 8
M2 4 16
M3 4 32
M4 8 16
Large Servers
L1 8 32
L2 8 64
L3 16 32
XL1 16 64
XL1 32 64
XL2 32 128

[0057] The big rule recommender: Classifies a plurality of migration approaches into one of a BigRule violation and a BigRule exceptions, providing checkpoint to finalize the migration approach recommended by the treatment type recommender and helps to plan & schedule the migration accordingly.
[0058] The above data is gathered and converted into Pandas data frame (data format for building a machine-learning model) using python script and Pandas library. Above prepared data is fed to the Decision tree model for generalization, estimations are calculated, and predictions will be generated with probability scores for each of the predictions. This ML model uses a two-model approach. One model will predict the Big Rule Violation and the Big Rule exception is predicted by the second model with the same input parameters.
[0059] During training and inferencing of the plurality of pretrained ML model, the segregated plurality of parameters relevant to the big rule recommender comprising Host Name Server Type Server_Role Application Name Application Group Part of Cluster (Yes/No) High Availability (Yes/No) CPU Total Cores > 32 Cores Memory > 128 GB Storage > 500 GB Database (Yes/No) Shared_LUN (Yes/No) Shared_Cluster (Yes/No);
[0060] All the inferences generated for BigRule recommendations are updated directly on the Master Discovery update section of the Infrastructure Migration Cockpit tool as mentioned in Table 3.
Table 3:
BigRule Violation BigRule Exception
Yes No

[0061] The move group recommender: Recommends a move group among a plurality of move groups, wherein the move group decides the migration event based applications, subnet or environment types that can migrate together to a target environment.
[0062] The identification of move groups is mandatory to generate migration event based applications or subnet or environment types that can migrate together to the target environment. Move group is logical grouping of Configuration Item (CI) based on application group, environment, or subnet. For example, CI part of the same application should logically be grouped together into move Groups for transformation and migration within a defined timeline.
[0063] More than one Move Group may run in parallel, and a Move Group may start after the completion of another.
[0064] This solution aims to automate the manual task of recommending the move groups using AI and update the output directly within the Infrastructure Migration Cockpit tool to finalize the move groups.
[0065] During training and inferencing of the plurality of pretrained ML model, the segregated plurality of parameters relevant to the move group recommendations comprising Host Name Server Type Role Application Name Application Group Subnet Environment Treatment Type Project Id
[0066] Data gathered from the above step in form of report is converted into a Pandas data frame (data format for building a machine learning model) using a python script and Pandas library for the next step i.e., building the Move Group Recommendation model. Above prepared data is fed to the Decision tree model for generalization, estimations are calculated, and predictions will be generated with probability scores for each of the predictions. Below block diagram explains the implementation of model training part and inference and storing part of solution.
[0067] All the inferences generated for Move group recommendations are updated directly on the Master Discovery update section of the Infrastructure Migration Cockpit tool as mentioned in Table 4 below.
Table 4
Application based move group: IAM_MG
Application +Environment based move group : IAM_Prod_MG
Application + Subnet based move group : IAM_192.168.0.0_MG

[0068] The event planner-scheduler: Recommends i) the planning and scheduling of one or more events of the migration event, scale out and scale down based on available bandwidth of resouces based on replication appliance available, ii) for a recommended number of servers further recommends resources, duration, weekly schedule, and monthly schedule, and iii) optimization of upcoming one or more events based on event analysis to increase velocity and success rate of the migration event.
[0069] This component aims to perform the below mentioned tasks using AI:
Based on history of something done, taking these data, recommend planning & schedule events, scale out & down/ based on available bandwidth, based on replication appliance available for 100 servers, how many resources, duration, weekly schedule, monthly schedule
Event analysis - Analysis of events to get recommendations to optimize upcoming migration events and increase velocity and success rate
First, step to use the data available in the master discovery section and history of updates for previous events in the Infrastructure Migration Cockpit tool. This data consists of the following input parameters.
[0070] There are two different inputs one set of inputs coming from the move group and event Id generation sections and another set of inputs coming from history of previous event updates. During training and inferencing of the plurality of pretrained ML model, the segregated plurality of parameters relevant to the
a) the event planner-scheduler comprising:
Move Group ID Move Group Name Event Id Event Name Total number CI Application Name Application Group Total Storage Size Migration Link Type Migration Bandwidth Migration Tool Replication Type Transfer Type Bulk Migration Overall Start Date Migration Start Date Initial Replication Start Date Initial Replication End Date Cutover Start Date Cutover End Date Overall End Date Total FTE Source Site Target Site Project Id Migration Quality Gates

[0071] Data Preparation: Data gathered from the above step in form of report is converted into a Pandas data frame (data format for building a machine learning model) using a python script and Pandas library for the next step i.e., building the Intelligent Plan and Schedule Recommendation model.
[0072] This component is implemented using three ML models for predicting Migration Bandwidth, Storage Size, Overall end date each. For this, the above prepared data is fed to the Decision tree model for generalization, estimations are calculated, and predictions will be generated with probability scores for each of the predictions.
[0073] ML model parameters, sample values and significance for different ML models is provided below:

Table 5: Treatment Type:
Parameters Name Value (Sample) Importance Significance of the parameter
Server_Type Physical High Server type is required to decide the migration tool and migration pattern
Server_Role Database High Server role is required to decide the criticality of application
Operating_System Windows Medium OS type is required to decide whether its compatible with target site or not.
CPU_Total_Cores 8 Medium CPU count is required to decide the resource availability in target site
Memory_RAM_GB 16 Medium Memory size is required to decide the resource availability in target site
Avg_CPU_Utilization_3months 5 Medium CPU utilization is required for right sizing
Avg_Memory_Utilization_3Months 12 Medium Memory utilization is required for right sizing
Total_Allocated_Capacity_GB 500 Medium Allocated capacity is required to decide the resource availability in target site
Total_Used_Capacity_GB 300 Medium Used Capacity is required to decide the resource availability in target site
LUN 100,101,102 High LUN number is required to find dependency with external storage
Storage_Name EMC Low Storage name is required to know storage vendor
Server_ISCSI iqn.yyyy-mm.naming-authority:unique High iSCSI is required to decide the storage transport protocol the LUN mapping on ESXi server
Server_WWPN No High WWPN is required to decide the storage transport protocol and LUN mapping on the ESXi server
No_of_NICs 1 Medium NIC count is required to validate the NIC requirement in target site
Part_of_Cluster Yes High Cluster type is required to decide the additional steps during migration execution, migration pattern and migration tool
Shared_LUN 100,101,102 High Shared LUN is required to decide the additional steps during migration execution, migration pattern and migration tool
Shared_Cluster Yes High Shared Cluster is required to decide the additional steps during migration execution, migration pattern and migration tool
NAS_Storage No High NAS storage is required to decide the migration execution, migration pattern, migration tool.
Mount_Location No High Mount location is required to decide the migration execution, migration pattern, migration tool.

Table 6: Move Group:
Parameters Name Value (Sample) Importance Significance of the parameter
Host Name tcsaisvcm01.tcs.local High Host name is generic
Server Type virtual High Server type is required to decide the migration tool and migration pattern
Role vCenter Appliance High Server role is required to decide the multi-tier or single tier application and criticality
Application Name VMware High Application name is required to group the servers belong to same application
Application Group Infrastructure High Application group is required to group the servers belong to same application group
Subnet 255.255.255.192 Medium Subnet is required to decide whether all the servers belong to single or multiple subnet
Environment Production High Environment is required to decide the environment type for the application and criticality
Treatment Type Virtual to Virtual Medium Treatment type is required to decide the migration tool, migration pattern and migration execution
Project Id XYZ001 Low Project id is generic
Table 7: Plan & Schedule:
Parameters Name Value (Sample) Importance Significance of the parameter
Move Group ID MG001 High Follow unique naming standard for reporting and governance perspective
Move Group Name MGTESTAPP01 High Follow unique naming standard for reporting and governance perspective
Event Id EVENT001 High Follow unique naming standard for reporting and governance perspective
Event Name EVENTTESTAPP01 High Follow unique naming standard for reporting and governance perspective
Total number CI 11 Medium Follow unique naming standard for reporting and governance perspective
Application Name VMware High Application name is required to group the servers belong to same application
Application Group Infrastructure High Application group is required to group the servers belong to same application group
Total Storage Size 500 GB High Storage size is required to decide the bandwidth required for migration
Migration Link Type Site to Site VPN High Migration link is required to check secured or non-secured or intranet or internet
Migration Bandwidth 10 Gbps High Bandwidth is required to calculate the timeline required to migrate maximum servers
Migration Tool Zerto High Migration tool is required to match the migration pattern or treatment type
Replication Type Hypervisor Based High Replication type is required to match the migration pattern or treatment type
Transfer Type Block High Transfer type is required to choose the migration tool
Bulk Migration Yes High Bulk migration is required to choose the migration tool
Overall Start Date Day/Month/Year High This is required to decide the migration timeline
Migration Start Date Day/Month/Year High This is required to decide the migration timeline
Initial Replication Start Date Day/Month/Year High This is required to decide the migration timeline
Initial Replication End Date Day/Month/Year High This is required to decide the migration timeline
Cutover Start Date Day/Month/Year High This is required to decide the migration timeline
Cutover End Date Day/Month/Year High This is required to decide the migration timeline
Overall End Date Day/Month/Year High This is required to decide the migration timeline
Total FTE 5 High This is required to decide the migration timeline and FTE requirement
Source Site Gujarat Medium Generic
Target Site Chennai Medium Generic
Project Id XYZ001 Low Generic
Migration Quality Gates Discovery, Plan, Migrate, Test and Handover High This is required to decide the migration timeline

[0074] At step 212 of the method 200, the one or more hardware processors 104 are configured to trigger a migration tool for the migration event to be initiated for migration of the datacenter to the target environment in accordance with the set of recommendations of each of the plurality of pretrained ML models.
[0075] FIG. 3 depicts a detailed process workflow of the system, in accordance with some embodiments of the present disclosure. As depicted in example implementation of the system of FIG. 3, the system 100 uses STICE™ along with these the ML based recommendations use machine learning algorithm such as random forest algorithm for classification of data and automatically generate recommendations. It is the most flexible and widely used algorithm as it can be used for both regression and classification. Random forest consists of multiple decision trees and combines them together to give more accurate and precise predictions. STICE can provide these required input parameters, and the ML models can be built model on top of this to provide analysis and generate these recommendations to improve planning and implementation of migration governance.
[0076] The system 100 in example of FIG. 3 uses:
[0077] Autonomous Discovery – An in house tool to discover the data and generate master discovery output.
[0078] STICE – “Discovery to Decommission” planning and implementation governance tool.
[0079] Jupyter notebook for Python Development and ML Model – Training, Cross-Validation and Testing.
[0080] Use autonomous discovery tool to collect the master discovery output and ingest these data to STICE™ discovery section. The same discovered data can be converted into easily readable by machine learning algorithm using python scripts. The ML model ( also referred to as Artificial Intelligence (AI) model) integrated to STICE™ is trained to predict the above-mentioned recommendations. Many inputs are coming from discovery output, which is related to individual CI, and very few inputs are coming from STICE™ system itself to train the AI model.
[0081] FIG. 4 depicts a Deep Learning (DL) model for treatment type recommendations, in accordance with some embodiments of the present disclosure. Deep Learning Model for training R3 Recommendation model or Treatment type recommendation – this model is trained from scratch without using inbuilt libraries. The model structure consists of 6 neural net layers wherein it has 1 input layer, 4 hidden layers and 1 output layer. The raw input taken from first layer for every neuron is multiplied with the weights and biases are added which was uniquely defined within the model result is then fed into the activation function. This activation function used was SoftMax that served with the non-linearity needed for the model to train its feedforward propagation cycle to determine the multiclass classification needed for the output layer.
[0082] FIG. 5 depicts a Decision tree and a random forest combined using ensemble method for training t-shirt sizing and defining move group recommendations, in accordance with some embodiments of the present disclosure.
[0083] Decision tree and random forest combined using ensemble method for training t-shirt sizing and defining move group recommendations – This is a hybrid model approach wherein both models are trained on same dataset with different algorithms and combined with the help of ensemble methods for getting better optimum predictions. The training dataset having raw data is scaled and transformed using feature engineering methods. For our multiclass classification solution Decision tree classifier and random forest are the two sets of classifier models that were chosen. Both algorithms give predictive values based on series of if-else conditions, these algorithms were fused together using BAGGing (bootstrapping + aggregation) an ensemble method approach which clubbed together those two classifiers to form one of the most efficient predictor and a unique ensemble model.
[0084] Feature engineering For Driving Reliable data – Besides using the standardized techniques for scaling and transforming the raw industry data this solution practices customizing inputs and exporting best suited features based on the experiences from the industrial experts and have built-in libraries for the same. Some of the major techniques practiced while transforming the data are Discretization (grouping and sorting), Imputation (handling missing values), feature splitting, handling outliers, Categorical encoding (one hot encoding), Variable transformation techniques that help with normalizing skewed data, scaling, and use of correlation matrixes.
[0085] The written description describes the subject matter herein to enable any person skilled in the art to make and use the embodiments. The scope of the subject matter embodiments is defined by the claims and may include other modifications that occur to those skilled in the art. Such other modifications are intended to be within the scope of the claims if they have similar elements that do not differ from the literal language of the claims or if they include equivalent elements with insubstantial differences from the literal language of the claims.
[0086] It is to be understood that the scope of the protection is extended to such a program and in addition to a computer-readable means having a message therein; such computer-readable storage means contain program-code means for implementation of one or more steps of the method, when the program runs on a server or mobile device or any suitable programmable device. The hardware device can be any kind of device which can be programmed including e.g., any kind of computer like a server or a personal computer, or the like, or any combination thereof. The device may also include means which could be e.g., hardware means like e.g., an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or a combination of hardware and software means, e.g., an ASIC and an FPGA, or at least one microprocessor and at least one memory with software processing components located therein. Thus, the means can include both hardware means, and software means. The method embodiments described herein could be implemented in hardware and software. The device may also include software means. Alternatively, the embodiments may be implemented on different hardware devices, e.g., using a plurality of CPUs.
[0087] The embodiments herein can comprise hardware and software elements. The embodiments that are implemented in software include but are not limited to, firmware, resident software, microcode, etc. The functions performed by various components described herein may be implemented in other components or combinations of other components. For the purposes of this description, a computer-usable or computer readable medium can be any apparatus that can comprise, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device.
[0088] The illustrated steps are set out to explain the exemplary embodiments shown, and it should be anticipated that ongoing technological development will change the manner in which particular functions are performed. These examples are presented herein for purposes of illustration, and not limitation. Further, the boundaries of the functional building blocks have been arbitrarily defined herein for the convenience of the description. Alternative boundaries can be defined so long as the specified functions and relationships thereof are appropriately performed. Alternatives (including equivalents, extensions, variations, deviations, etc., of those described herein) will be apparent to persons skilled in the relevant art(s) based on the teachings contained herein. Such alternatives fall within the scope of the disclosed embodiments. Also, the words “comprising,” “having,” “containing,” and “including,” and other similar forms are intended to be equivalent in meaning and be open ended in that an item or items following any one of these words is not meant to be an exhaustive listing of such item or items or meant to be limited to only the listed item or items. It must also be noted that as used herein and in the appended claims, the singular forms “a,” “an,” and “the” include plural references unless the context clearly dictates otherwise.
[0089] Furthermore, one or more computer-readable storage media may be utilized in implementing embodiments consistent with the present disclosure. A computer-readable storage medium refers to any type of physical memory on which information or data readable by a processor may be stored. Thus, a computer-readable storage medium may store instructions for execution by one or more processors, including instructions for causing the processor(s) to perform steps or stages consistent with the embodiments described herein. The term “computer-readable medium” should be understood to include tangible items and exclude carrier waves and transient signals, i.e., be non-transitory. Examples include random access memory (RAM), read-only memory (ROM), volatile memory, nonvolatile memory, hard drives, CD ROMs, DVDs, flash drives, disks, and any other known physical storage media.
[0090] It is intended that the disclosure and examples be considered as exemplary only, with a true scope of disclosed embodiments being indicated by the following claims.
, Claims:We Claim:
1. A processor implemented method (200) for datacenter migration, the method comprising:
performing (202), by one or more hardware processors, data discovery from a master Configuration Item (CI) list, a master discovery list, event updates and bulk migration updates, wherein a discovered data comprises a plurality of parameters associated with an existing infrastructure of a datacenter;
pre-processing (204), by the one or more hardware processors, the plurality of parameters;
segregating (206), by the one or more hardware processors, the pre-processed plurality of parameters into a plurality of groups, wherein each group among the plurality of groups is associated with a pretrained Machine Learning (ML) model among a plurality of pretrained ML models based on a predefined relevancy of each of the preprocessed plurality of parameters with each of the plurality of pretrained ML models the plurality of pretrained ML models comprising a treatment type recommender, a T-shirt sizing recommender, a big rule recommender, a move group recommender, And an event planner-scheduler;
extracting (208), by the one or more hardware processors, a set of optimum features, for each of the plurality of pretrained ML models, from the segregated plurality of parameters of the group based on a reference library and domain knowledge;
generating (210), by the one or more hardware processors, for a migration event of the datacenter towards a target environment, a set of recommendations by each of the plurality of pretrained ML models based on the set of features extracted for each of the plurality of pretrained model, wherein,
the treatment type recommender recommends a pair of treatment types and associated pattern indicating a migration approach to be used,
the T-shirt sizing recommender bundles a fixed amount of RAM and storage with a given number of virtual CPU cores based on need for scale-up or scale-down of compute resources post migration,
the big rule recommender classifies a plurality of migration approaches into one of a BigRule violation and a BigRule exceptions, providing checkpoint to finalize the migration approach recommended by the treatment type recommender that enables to plan and schedule the migration accordingly
the move group recommender recommends a move group among a plurality of move groups, wherein the move group decides the migration event based applications, subnet or environment types that can migrate together to a target environment, and
the event planner-scheduler recommends i) the planning and scheduling of one or more events of the migration event, scale out and scale down based on available bandwidth of resources, based on replication appliance available, ii) for a recommended number of servers further recommends resources, duration, weekly schedule, and monthly schedule, and iii) optimization of upcoming one or more events based on event analysis to increase velocity and success rate of the migration event; and
triggering (212), by the one or more hardware processors, a migration tool for the migration event to be initiated for migration of the datacenter to the target environment in accordance with the set of recommendations of each of the plurality of pretrained ML models.

2. The method as claimed in claim 1, wherein during training and inferencing of the plurality of pretrained ML model, the segregated plurality of parameters :
a) for the treatment type recommender comprise server_Type, Server_Role, Operating_System, CPU_Total_Cores, Avg_CPU_Utilization_3months, Memory_RAM_GB, Avg_Memory_Utilization_3Months Total_Allocated_Capacity_GB, Total_Used_Capacity_GB, LUN, Storage_Name Server_ISCSI, Server_WWPN, No_of_NICs, Part_of_Cluster Shared_LUN, Shared_Cluster, NAS_Storage, and Mount_Location,
b) for the T-shirt sizing recommender comprise Server_Type Server_Role Operating_System CPU_Total_Cores Avg_CPU_Utilization_3months Memory_RAM_GB Avg_Memory_Utilization_3Months Total_Allocated_Capacity_GB Total_Used_Capacity_GB LUN Storage_Name Server_ISCSI Server_WWPN No_of_NICs Part_of_Cluster Shared_LUN Shared_Cluster NAS_Storage Mount_Location,
c) for the big rule recommender comprise Host Name Server Type Server_Role Application Name Application Group Part of Cluster (Yes/No) High Availability (Yes/No) CPU Total Cores > 32 Cores Memory > 128 GB Storage > 500 GB Database (Yes/No) Shared_LUN (Yes/No) Shared_Cluster (Yes/No),
d) for the move group recommendations comprise Host Name Server Type Role Application Name Application Group Subnet Environment Treatment Type Project Id, and
e) for the event planner-scheduler, comprise:
Move Group ID Move Group Name Event Id Event Name Total number CI Application Name Application Group Total Storage Size Migration Link Type Migration Bandwidth Migration Tool Replication Type Transfer Type Bulk Migration Overall Start Date Migration Start Date Initial Replication Start Date Initial Replication End Date Cutover Start Date Cutover End Date Overall End Date Total FTE Source Site Target Site Project Id Migration Quality Gates.

3. The method as claimed in claim 1, wherein the plurality of migration approaches comprise i) Physical to Physical, Re-host ii) Physical to Virtual, Re-host iii) Virtual to Virtual, Re-host iv) Lift and Shift, Lift and Shift v) Data Migration, Build and Restore vi) Storage Area Network (SAN) Migration, Re-host/ Lift and shift, and vii) Network Attached Storage (NAS) Migration, Build and Restore.

4. The method as claimed in claim 1, wherein the preprocessing comprises discretization, imputation, feature splitting, handling outliers, categorical encoding, scaling, and correlation matrixes.

5. A system (100) for data center migration, the system (100) comprising:
a memory (102) storing instructions;
one or more Input/Output (I/O) interfaces (106); and
one or more hardware processors (104) coupled to the memory (102) via the one or more I/O interfaces (106), wherein the one or more hardware processors (104) are configured by the instructions to:
perform data discovery from a master Configuration Item (CI) list, a master discovery list, event updates and bulk migration updates, wherein a discovered data comprises a plurality of parameters associated with an existing infrastructure of a datacenter;
pre-process the plurality of parameters;
segregate the pre-processed plurality of parameters into a plurality of groups, wherein each group among the plurality of groups is associated with a pretrained Machine Learning (ML) model among a plurality of pretrained ML models based on a predefined relevancy of each of the preprocessed plurality of parameters with each of the plurality of pretrained ML models the plurality of pretrained ML models comprising a treatment type recommender, a T-shirt sizing recommender, a big rule recommender, a move group recommender, And an event planner-scheduler;
extract a set of optimum features, for each of the plurality of pretrained ML models, from the segregated plurality of parameters of the group based on a reference library and domain knowledge;
generate for a migration event of the datacenter towards a target environment, a set of recommendations by each of the plurality of pretrained ML models based on the set of features extracted for each of the plurality of pretrained model, wherein,
the treatment type recommender recommends a pair of treatment types and associated pattern indicating a migration approach to be used,
the T-shirt sizing recommender bundles a fixed amount of RAM and storage with a given number of virtual CPU cores based on need for scale-up or scale-down of compute resources post migration,
the big rule recommender classifies a plurality of migration approaches into one of a BigRule violation and a BigRule exceptions, providing checkpoint to finalize the migration approach recommended by the treatment type recommender that enables to plan and schedule the migration accordingly
the move group recommender recommends a move group among a plurality of move groups, wherein the move group decides the migration event based applications, subnet or environment types that can migrate together to a target environment, and
the event planner-scheduler recommends i) the planning and scheduling of one or more events of the migration event, scale out and scale down based on available bandwidth of resources, based on replication appliance available, ii) for a recommended number of servers further recommends resources, duration, weekly schedule, and monthly schedule, and iii) optimization of upcoming one or more events based on event analysis to increase velocity and success rate of the migration event; and
trigger a migration tool for the migration event to be initiated for migration of the datacenter to the target environment in accordance with the set of recommendations of each of the plurality of pretrained ML models.

6. The system as claimed in claim 5, wherein during training and inferencing of the plurality of pretrained ML model, the segregated plurality of parameters :
a) for the treatment type recommender comprise server_Type, Server_Role, Operating_System, CPU_Total_Cores, Avg_CPU_Utilization_3months, Memory_RAM_GB, Avg_Memory_Utilization_3Months Total_Allocated_Capacity_GB, Total_Used_Capacity_GB, LUN, Storage_Name Server_ISCSI, Server_WWPN, No_of_NICs, Part_of_Cluster Shared_LUN, Shared_Cluster, NAS_Storage, and Mount_Location,
b) for the T-shirt sizing recommender comprise Server_Type Server_Role Operating_System CPU_Total_Cores Avg_CPU_Utilization_3months Memory_RAM_GB Avg_Memory_Utilization_3Months Total_Allocated_Capacity_GB Total_Used_Capacity_GB LUN Storage_Name Server_ISCSI Server_WWPN No_of_NICs Part_of_Cluster Shared_LUN Shared_Cluster NAS_Storage Mount_Location,
c) for the big rule recommender comprise Host Name Server Type Server_Role Application Name Application Group Part of Cluster (Yes/No) High Availability (Yes/No) CPU Total Cores > 32 Cores Memory > 128 GB Storage > 500 GB Database (Yes/No) Shared_LUN (Yes/No) Shared_Cluster (Yes/No),
d) for the move group recommendations comprise Host Name Server Type Role Application Name Application Group Subnet Environment Treatment Type Project Id, and
e) for the event planner-scheduler, comprise:
Move Group ID Move Group Name Event Id Event Name Total number CI Application Name Application Group Total Storage Size Migration Link Type Migration Bandwidth Migration Tool Replication Type Transfer Type Bulk Migration Overall Start Date Migration Start Date Initial Replication Start Date Initial Replication End Date Cutover Start Date Cutover End Date Overall End Date Total FTE Source Site Target Site Project Id Migration Quality Gates.

7. The system as claimed in claim 5, wherein the plurality of migration approaches comprise i) Physical to Physical, Re-host ii) Physical to Virtual, Re-host iii) Virtual to Virtual, Re-host iv) Lift and Shift, Lift and Shift v) Data Migration, Build and Restore vi) Storage Area Network (SAN) Migration, Re-host/ Lift and shift, and vii) Network Attached Storage (NAS) Migration, Build and Restore.

8. The system as claimed in claim 5, wherein the preprocessing comprises discretization, imputation, feature splitting, handling outliers, categorical encoding, scaling, and correlation matrixes.

Dated this 25th Day of August 2023
Tata Consultancy Services Limited
By their Agent & Attorney

(Adheesh Nargolkar)
of Khaitan & Co
Reg No IN-PA-1086