Blogs

Migrating to iSCSI storage? We have you covered!

By Ravi Prakash, Director Product Marketing

When you migrate to iSCSI what should you be thinking about?

  • What workloads are you planning to migrate from Fibre Channel to iSCSI
  • How do you compare Fibre Channel workload performance to that of iSCSI?
  • How do you assure ongoing application performance and capacity once you migrate to iSCSI?

Migrating to iSCSI

If you are planning to migrate your application workloads from Fibre Channel block storage to iSCSI storage, Virtual Instruments can help you on this journey!

Virtual Instruments is the leader in Application workload monitoring and validation from production to the lab.  Starting with the ability to monitor and characterize your application workloads in production with VirtualWisdom, to simulating and performing what-if scenarios in the lab with WorkloadWisdom.

WorkloadWisdom is designed to analyze, model and simulate your workloads based on your production storage data with the industry’s highest fidelity coming from the integration with VirtualWisdom.  You can view a workload in terms of read, write, metadata operations, block/file size distribution and more.

It is easy to replay your application workload behavior against either FC or iSCSI storage targets and compare them.   The WorkloadWisdom iteration engine is data driven and automates the testing of hundreds and thousands of permutations by iterating through a given set of parameter values.    This allows you to find breaking points or sweet spots of the iSCSI storage targets you are evaluating until you identify the storage setup which meets your workload requirements.

Migrating to iSCSI Workflow

Once you begin migrating to iSCSI how do you ensure that applications are meeting their SLAs to the business?  Read on for the answer…

Monitor iSCSI application conversations at wire-level

VirtualWisdom is an AIOps platform that provides a “single pane of glass” across your hybrid application infrastructure.  VirtualWisdom delivers proactive application-centric dashboards that provide recommendations to optimally balance your workloads in production and can be tailored for application owners, executives and many other roles within IT Operations.

Our dashboards are built on full-stack monitoring for your infrastructure from applications to data and VirtualWisdom is unique in its ability to combine compute, network and storage machine data with high fidelity wire data to measure every storage transaction in a Fibre Channel or iSCSI environment.

The VirtualWisdom purpose-built Performance Probe for NAS and iSCSI gives you visibility into every application conversation over iSCSI at the highest fidelity available in the industry.

Migrating to iSCSI - Dashboard for VP Infrastructure

Consider the dashboard above which could be used by a VP or Director of Infrastructure:

  • The first panel “Applications by Open Case” displays an open case associated with the Order Manager application which is an OLTP application running in 4 different virtual machines. This open case is focuses on iSCSI write response time.
  • The second panel “Top Applications by CPU Ready” shows data that is obtained using the software integration for vCenter.
  • The third panel “Top Applications by Network Throughput” shows data obtained using the software integration for NetFlow (NetFlow can be generated by the VMware vSphere Distributed Switch or from physical IP routers and switches).
  • The fourth panel “Top Applications by Storage Response Times” shows data obtained using the Performance Probe for NAS, which support the iSCSI protocol along with NFS and SMB.
  • The bottom-right panel displays recommendations which were automatically generated by VirtualWisdom’s VM Coordinator This purpose-built analytic uses Artificial Intelligence (AI) and Machine learning (Monte Carlo simulation) to learn VM workloads and recommend an optimal placement of VMs in an ESX cluster to proactively avoid memory, disk and/or CPU contention.

Let us investigate the Order Manager application in further detail to identify root cause of its inability to meet an agreed upon SLA between IT and the business.

Migrating to iSCSI - Order Manager Dashboard

Order Manager is a Platinum tier application, running in 4 virtual machines and using iSCSI block storage.  In the Storage Health panel there is a notification for an alarm on one of the front-end ports of the iSCSI storage system. Selecting this notification automatically takes you to the relevant case where you’re given additional details and guidance about the alarm:

Migrating to iSCSI - Payload Rate

From the chart above we observe that an alarm threshold has been breached which is impacting two applications: Order Manager (a Platinum tier application) and Supply Chain ETL (a Silver tier application). VirtualWisdom can automatedly discover application configuration and business criticality from multiple ITSM and APM solutions.

Using this information, we can ask Trend Matcher (another one of VirtualWisdom’s purpose-built analytics) to help use figure out what is impacting the Order Manager application.

Migrating to iSCSI - Trend Matcher

From the Trend Matcher results we conclude that Supply Chain ETL is behaving like a noisy neighbor and impacting the SLA of the mission critical Order Manager application!

This is just one of many issues you can resolve using VirtualWisdom, the data it collects (from a variety of software integrations), and the Performance Probe for NAS.

Note it is not only the ability to collect iSCSI metrics at high fidelity that makes VirtualWisdom unique. VirtualWisdom can ingest data for the entire infrastructure stack and correlate the resulting 17,000 or more unique metrics using Artificial Intelligence and Machine Learning-based analytics to help customers avoid application downtime and improve MTTR by providing automated recommendations and root cause analysis for applications within your infrastructure.

Like what you hear and wish to learn more? REQUEST DEMO NOW

Don’t forget to follow us on TwitterLinkedIn and Facebook to stay up-to-date on the latest and greatest in app-centric infrastructure performance monitoring.