Brahamaprakash Vardhanan

1) OS Certification Team lead:

Leading team of 6 members for OS Certification on Cisco UCS Rack and Blade server for tracking releases, delivery of the projects on time, Working with OS vendors for resolving issues related drivers, BIOS and firmware. Delivering the ESX, Redhat, Windows, Ubuntu, Xen, Oracle OS certification for each releases on the Cisco UCS Rack/Blade server, for the UCS hardware changes and new CPU releases.

Representing the team in the daily standup meeting, Tracking /scrubbing OS bugs across the team. And reproducing the customer testing scenarios with root causing the issues to closure.

Creating complex setups similar to data center environment with testing performance, disk latencies issues, for the different hypervisors like ESX, Citrix server.

2) Cisco UCS, UCSM Compat, interop, certification testing of OS,

Period: Sept to still date.

Team size: 2 and 6

Brief of project’s: The interop compt testing involves qualifying the cisco UCS, UCSM VIC card, LOM, RAID and flash card’s on different OS’s .

The work involves with qualifying the above cards with taking inbox/outbox driver, firmware and then installing the OS,’s with different types of installation like SAN,Iscsi and PXE.Then installing the OS on different types RAID levels on different types of Hard disk/SSD’s.Raising defects with development team for different types installation and errors in the OS involves as part of regression testing. Preparing the test matrix, for qualifying cisco, and third party adapters for different types of OS releases for different types of driver and then publishing into Interop tool of the cisco as part of Hardware compatibility list (HCL).

3) VSAN, Vflash qualification on cisco UCS server.

Team size: 3

Brief of project: One of the primary design goals for Virtual SAN is to deliver a new level of storage management simplicity and ease-of-use to the VMware administrator. Virtual SAN achieves this by creating an abstracted pool of storage resources that can be used easily and efficiently across the entire vSphere cluster.

The qualification involves setting HDD,SSD and then creating the single vSAN datastore for RAID, and JBOD mode of disk’s.The qualification involves for LSI controller level testing with running the lots of IO and checking any crash of the guest VM on ESXi server.

Testing different types disk like SSD,HDD combination of Hybrid and All flash configuration , ejections cluster on different types network and qualified disk with setting the vflash of different size for testing the read and write testing is the part of qualification of VSAN on cisco UCS server.

4)Nvidia Grid and Tesla mode GPU,VGpu Testing on ESXi/Citrix.

Team size : 2

Brief of Project: This is to qualify the GRID and Tesla cards from Nvidia on Cisco UCS server with testing the Pass trough and non pass-through GPU card’s.

The testing involves the installing driver of Nvidia on windows, Linux Guest VM of ESX,Citrix Hosts. Then testing the Stress test of the guest OS for longer duration with connecting from VMware View-client Xen receiver /XenDesktop for longer duration to observe an crash of the VM’s. Creating multiple vGPU profile for multiple VM’s and then testing VM state on ESXi and Citrix hosts was part of testing and publishing into cisco and Respected vendors HCL website.

Period: March 2011 to May 2012

Client: Netapp Bangalore

Team size: 3

Brief about Project: The main aim of this project is to certify they citrix Xen 6.0 GA, features on the Netapp’s ONTAP FILER exported Storage via NFS, where the Citrix Xen OS is being installed on the different hardware’s like Fujitsu blade , IBM blade, and Cisco UCS server.

The Qualification was done for the different version of Netapp FILER release’s like BR 8.0.1/2 ,IC 7.3.6 and 8.1RC2 on the both 7 mode and 10 Mode (Cluster ) of Filer with Citrix 6.0 GA release.

Roles performed:

• Created Test plan, Test configuration, and Test cases from the scratch.
• Created Test Matrix for Citrix and FILER release for different Schedules for the testing cycles.
• Configured the Hardware including setting up of connection from FILER to switch, and Switch to Blade servers (Hosts).
• Done Testing with configuring the network connection in the switch and FILER for the single, Multimode and Multi with LACP configurations to test the active-passive and active-active network IO connections respectively.
• Tested the feature of VIF failover, and NIC failover for high availability and the active IO is falling in another link.
• Configured the Cluster on FILER for 10 mode with creating the virtual server and network interfaces for the cluster ,and tested the scenarios of Takeover/Giveback,

Force takeover/force giveback, Panic, and unplanned shutdown of FILER with IO being continuously generating in the Xen server Guest Hosts.

• Tested the following features of Xen server like, XenMotion, Xen HA, work load balancing, NIC bonding VM protection. Policies.
• Tested the Hot and Cold migration for the Xenserver with migrating the guest VM’s during transfer of IO.
• Tested rigorously the features of snapshots of Xen server both from Xen Center and through XE command line interface.
• Mounted the NFS export from FILER and tested the NFSv3 and NFSv4 features like locking and delegation and mounting features for the different guest OS’s .
• Tested the HA features with configuring the restart policy of guest VM’s and heartbeat configurations which being hosted on NFS/ISCSI Storages.
• Tested the NDU (Non disruption upgrade) from the FILER side with upgrading from existing version to latest version, with keeping in mind IO’s won’t get disturbed, and network connections are intact during the upgradation process.
• Tested network throughput with enabling the jumbo frames, from both FILER and Citrix side and verified the same through wireshark network analyser.

2) NFS Interop Testing on ONTAP

Period: June 2010 to Jan 2011

Client: Netapp Bangalore

Team size: 3

Brief about Project: This project involves testing of NFS feature for ONTAP release version of IC and BR releases.

The basic Feature of NFS involves testing different kinds of export’s from FILER side and mounting the exported NFS file system in the clients with different options

Roles performed:

• Leading the team of 3 members to test the NFS on different releases of FILER.
• Created Test plan for the above NFS Interop testing.
• Written Test cases for the different types of NFS features like different types of mounting NFS file system for NFSV2, NFSV3, NFSV4 on Linux and Solaris clients..
• Executed the Automated scripts of NATE framework for the above cases and analyzed the logs for any failures.
• Tested NFS features like locking, delegations, Named Attributes, Multiprotocol, and CIFS.
• Tested the Load and performance testing of the NFS features for the different Interop clients to see the NFS Performance and make benchmark for the same.
• Analyzed the logs with colleting the Network logs and analyzing same in the wireshark for testing different types of protocols.
• The testing was carried on the different types of FILERS, C -Mode, 7- Mode and for different clients like Linux, Solaris, AIX,and HPUX for their different releases.

Project 1

Title: File Replication

Environnent: Windows (2k3, 2k, Xp), Linux (RHEL, Suse), UNIX (HP_UX, Solaris, AIX, MacOs)

Role: Team member, Lead

Brief about project

DR Scout FX File Replication allows asynchronous replication of complete file systems from the source machine to the target machine.

Basically the main purpose of this product is to take the backup of files in the asynchronous manner and make this as as continuous backup of files, by scheduling the backup with particular intervals of time.

Project 2

Title: SQL 2000 / SQL 20005 Reporting server Replication

Environment: Windows (2k3,2zk)

Role: Team member

To support reporting servers with a copy of an actual database DR-Scout uses VX agents to replicate the database to a target on which reporting server can operate on. Similarly reporting servers have their own meta-data that will be replicated to another target (or the same target).

Creating a copy of a volume from one location to another location. Thus replicated volumes are copied to a target volume, but unlike the synchronous case, DR Scout does not wait for the secondary storage to acknowledge the write before allowing primary storage to continue writing to disk.

Role Performed: Done following types of testing for the above mentioned project.

Project 3

Title: SQL 2000 / SQL 20005/SQL 2008 and Oracle 10g Failover/Failback

Environment: Windows (2k8 ,2k3,2k), Linux ,UNIX (AIX)

Role : Team member, Lead

Brief about project

SQL 2000/2005

Basically the aim of this Failover is to replicate the Database of SQL 2000/SQL 2005 ,Oracle to the target Machine from the source using our Vx agent (volume Replication).

And in the target the Db can be mounted with using our Vx solution.

Project 4

Title: Fabric Based Protection

Environment: Windows (2k3,2k), Linux, Windows 2k3 Cluster

Role : Team member

Brief about project

The InMage solution/Avalon0.1 is a disaster recovery solution for a Brocade intelligent fabric which provides continuous data protection and continuous remote replication to and from local and remote heterogeneous storage arrays.

Project 5

Title: Citrix Virtualization backup

Environment: Citrix Xen server 4.0/4.1/5.0

Role : Team member, lead

Brief about project

The project involves basically 2 Solutions, one with LV (Logical volume) backup and another SR(Storage Repository) backup solution.

The main aim of the project is taking the Citrix Server Backup where there may be n Number of Guest VM’s running under the parent host , so if user want to protect the particular Guest VM ,then he may opt for the LV solution, if they wants to protects whole of the disk where there are n number of VM’s are running under that Disk(either FC or iscsi exported one) then he may take backup of that whole Disk (SR in Citrix) .

Basically the SR solution was completely automated using the shell script and internal commands of Xen Citrix server.

Project 6

Title: Citrix Virtualization backup using Pool

Environment: Citrix Xen server 4.0/4.1/5.0

Role: Team member, lead

Brief about project

The idea behind this project is to protect the Guest VM’s running under the Xen Citrix Server where the VM’s can Failover or Failback having High availability function (HA).So we developed the option to protect the VM’s by just creating the respected VM as LV on our UI and just selecting the respected VM (LV) on UI and then replicating that LV to Target Xen Pool or Standalone server.

In Target we are equally checking the respected VM (LV) is replicated and can boot up without having any issues.

Project 7

Title: Solaris Volume Based protection

Environment: Solaris 8/9/10 sparc/x86

Role: Team member

Brief about project

The Basic aim of this project is to protect the host in the solairs platform, for the volume based protection.

The Volume’s of the Solaris partition may be either having full disk or may be the Logical volume manager’s. Partition.

The basic file system of volume used here is UNIX file system (UFS) and veritas file systems (VXFS) .And the volume manager used to create Logic volumes are by using Sun volume manager or Veritas volume manager.

The process involves of having selecting source based Volume’s having any of the file systems (UFS or vxfs) and then selecting the target having raw volume (created by select disk partitions).

Once the replication process is over, taking snapshot with recovering the target volumes with many point in time’s or through issue consistency tag with freezing the file systems.

The testing involved of covering in different platforms like Solaris 8/9/10 and having pool environment by using zpool and Solaris containers in solaris 10 as base.

The Volume managers with Sun are created by using Soft partitions by using the different RAID groups (preferably RAID 0,RAID 1 and RIAD 5).

) Project Name: Automation of USD CLI, using Dmscripting.

This project is implemented for testing a functionality as well as performance of the product called Unicenter Software Delivery (USD).

The USD basically used for deploying the software packages and patches etc to different agent machines (client) from an Enterprise server (Server).

The DMScripts (Desktop Management) an scripting language is used for CA’s internal testing, which is also supported of Command Line Interface (CLI) utility A Dmscript debugger, which will, comes After Installing a USD in windows for executing scripts and can be made to execute the scripts by writing an shell script in Linux platform.

Work done on this feature:

· Written Test plan for this feature.

· Done and System Testing for this product

· Involved in writing Test Scripts by using the USD commands for building the Automation.

· Involved for writing Library Functions for the Framework of automation by using the Dmscripting Language.

· Done a Performance, Sanity and Regression Testing.

· Made the Whole Testuite to run on different managers (servers) like Enterprise and Domain manager by connecting different agents (clients)

2) USD Automation of CLI in ITRM DSM product.

This project is continuation of the first project, where in which we are testing same functionality of USD CLI, but for a product called Information Technology Resource Management (ITRM), Desktop and server Management (DSM). Since the product involves an integration of 3 major products called USD(Unicenter Software Delivery) UAM (Unicenter Asset management) and URC (Unicenter Remote control) which went many massive changes like issues of migration, security etc.

Work done on this feature:

· Written Test plan for this feature

· Performed an System Testing for this product

· Performed an Acceptance, and ad-hoc testing.

· Raised many severity 1 and 2 issues for finding bugs, done bug verification.

· Involved in writing Test Scripts by using the USD commands for building the Automation.

· Done a Performance and Sanity and Regression Testing.

3) ITRM R11 Performance and Scalability testing by using Load tester.

This project is basically used for measuring performance of the product of ITRM (Information Technology Resource Management) by using one of the internal tool developed called Laodtester,

This Load tester provides the result of the testing how many messages are passed /failed. The replication of the messages may involve the 3 product called Unicenter Remote Control (URC), Unicenter Software Delivery (USD) and Unicenter Assett management).

Work done on this feature:

· Written Test plan for this feature

· Done a Performance and Sanity Testing

· Performed an Acceptance, and ad-hoc and Regression Testing

· Raised Issues for performance and did bug verification.

· Wrote a batch file and scripts to run the load tester.