click to expand ...

Home
About SVA ... Contact us
Overview (all documents)
Product description
Services with BVQ
BVQ advantages
New functions
White papers
Downloads and releases
Users Manual and users guide
Further WIKI documents
Performance analysis and monitoring for IBM SVC and Storwize

 



The Challenge

Business’s that run their most critical applications on VMware and store their data on IBM Spectrum Virtualize Storage have a challenge; to meet the performance requirements of a variety of production

applications and users.  Administrators need to make sure these performance requirements are met on a consistent basis.

An efficient way of monitoring and resolving performing issues across both environments requires an insight that includes a deep integration between both environments where the performance is correlated between the two layers.

In addition, having the ability to look across all VMs spanning across several ESXi Hosts, makes it easier to diagnose a performance issue within minutes.

When this insight and flexibility into both layers is not available to the administrator, finding the slow drainer in the VMware environment can be difficult and can take away time from other critical duties.

True end-to-end monitoring and performance analysis across VMware and Storage

BVQ is the solution that offers true end-to-end monitoring and performance problem resolution of the VMware and Spectrum Virtualize Storage environments.  BVQ offers a comprehensive monitoring and troubleshooting solution to monitor both environments efficiently from a unified interface with a correlation of data across both VMware and Storage environments.  In this way, helping both VM and storage administrators perform their day to day activities more efficiently.

Here we show you how when using BVQ, you have the ability to go from one environment to the other very seamlessly in order to identify performance issues right away without having to navigate through many views to identify where the issue is coming from.


Use case: BVQ helps determine where the performance is coming from in the VMware environment

Users are experiencing a performance degradation on a production application.  BVQ offers these functionalities to efficiently find where the performance issue is occurring:

  • Monitor and see activity levels of a large number of VMs at once from a single view

  • BVQ can see all VMs spanning across several ESXi Hosts, making it easier to diagnose a performance issue within minutes

  • The storage or VM administrator can pinpoint quickly if the performance issue impacting the user is coming from the VM resource saturation or a storage performance issue

 BVQ can see all VMs spanning across several ESXi Hosts, making it easier to diagnose a performance issue within minutes.

With the favorite "VMWare Performance" you can see the activity levels of a large number of VMs at once from a single view - not only that, but see all VMs spanning across several ESXi Hosts. 

  • Easily identify a saturation in the memory or CPU on the ESXi host and on the particular VM causing the saturation. 

    Then remediation steps can be implemented to either move VMs to different hosts or add more resources to the host.

  • See the IO load on the VM LUN -data rate and Latency to check for spikes. From identified spikes that surpass threshold level, determine the issue is coming from the storage side and  identify which VM is causing that peak load.





The storage admin can pinpoint quickly if the performance issue impacting the user is coming from the VM resource saturation or a storage performance issue:could be several hundred VMs at once from a single view and  and identify where it’s coming from, i
BVQ provides a performance correlation between the two layers

BVQ wil help you

-Which VM is driving that particular IO load? Yes because it can see al VMs running on this backend volume of the ESX host. And pinpoint the particular VM driving the high IO load. You have aidentified the bottleneck,now this vm can be moved to a different host or dig further

-Can drill down to the not only to the Datastore level , but it can also go to the backend  to the SCSI lun.  Let me show you how, show favorite

 

Here you can see which ESX is causing a high load on the backend disk.

 

Also can identify which Vm is driving that particular high IO load (here)and go to the SCSi LUN vm is using

-and see the performance Io and latency of the virtual disk

-and see the volume on the storage that it is using. And immediately see if there is a latency issue with the volume so that further analysis can be donwn ow on the storage side

 




Steps to solve the problem

  1. It has to be found out why the storage responds to the higher load with this high latency. This is a standard storage analysis supported by BVQ.
  2. If the storage is overloaded, a decision must be made whether to move this virtual machine to another storage.
  3. If a SAN problem is detected, you can do a SAN analysis with the BVQ SAN extension.

 

  • No labels