Is All-NVMe Worth Your Money? – WorkloadWisdom 6.2 Virtual Instruments Briefing Note

All-Flash Arrays (AFA), the goto high performance storage system for the past few years, is being replaced by All-NVMe. NVMe flash promises to improve performance by reducing latency and increasing bandwidth to flash based media. NVMe is PCIe based instead of SAS based and it supports more simultaneous commands and a deeper queue depth. The protocol is also networkable with NVMe over Fabrics support available for both Fibre Channel and Ethernet networks.

The reality is that many data centers are not pushing the limits of their current Fibre and Ethernet attached SAS-based all-flash arrays. Additionally, other data centers find hybrid systems meet their performance needs. The problem is that most data centers simply don’t know the exact performance requirements of their applications and workloads. This lack of knowledge forces organizations to guess at their performance requirements. They often guess high, which means they pay a premium for performance they don’t likely need – leading to wasted spending.

Storage Switzerland is a proponent of organizations establishing a storage performance validation practice which continuously verifies the performance requirements of applications and workloads in the environment. It should then test these requirements against current and potentially new storage systems to align purchases and deployments to real-world performance requirements.

The challenge is how to assemble a lab capable of recreating the production environment and pushing new storage systems like All-NVMe arrays. These arrays claim millions of IOPS and most data centers don’t have the budget to set aside a lab environment to validate those claims. IT administrators are often forced to test workloads in production, risking application availability and performance.

Introducing Virtual Instruments WorkloadWisdom 6.2

Virtual Instrument’s WorkloadWisdom is a comprehensive storage workload performance validation and testing solution. It works by first analyzing and modeling production workloads. It then can replay those workloads on other storage systems that the organization either already owns or potentially new systems it is considering purchasing. It can also validate changes in the environment by comparing historical models.

WorkloadWisdom enables organizations to validate workloads in a lab, not in production. It creates a lifecycle approach to validating the organization’s infrastructure with real workload behavior. IT no longer needs to cobble together lab equipment in an attempt to simulate production; WorkloadWisdom replays the actual production workload model on lab storage systems or vendor evaluation units. It reduces the test environment to a workload generation appliance, with the storage network and the storage system going through validation/evaluation.

WorkloadWisdom is able to test a variety of protocols and storage types. For connectivity it supports 16GFC, 32GFC, 10GE, 25GE, 40GE. For storage targets it supports block (FC-SCSI, iSCSI, OpenStack and Cinder), File (SMB and NFS) and Object (S3, OpenStack, Swift, CDMI). The headline feature of the new 6.2 release is its support of NVMe over FC, enabling testing of the next generation of storage systems and networks.

Also in the 6.2 release Virtual Instruments improves its support for large scale test environments. The release includes a FC port browser, iSCSI auto-discovery and SMB/NFS bulk test user management. The 6.2 release also includes data integrity verification. With a single click, IT testers can now verify that data received back from the device being validated is the same as was written. Lastly, the new release includes real-time alerting to flag failures as they occur during testing. These alerts combine with new diagnostics to pinpoint command and data failures during tests.

StorageSwiss Take

Data centers should take a “trust but verify” approach to vendor claims, especially in the NVMe era. The problem is the cost and time required to assemble and manage a testing lab to verify claims is exceeding the budget capabilities of most organizations. Alternative approaches like Virtual Instruments’ WorkloadWisdom enable organizations to not only lower the cost to test new storage systems but also increase the accuracy of those tests.

To learn more about verifying if NVMe is right for your data center join Storage Switzerland and Virtual Instruments for our on demand webinar “Does Your Data Center Need NVMe?”.

Sign up for our Newsletter. Get updates on our latest articles and webinars, plus EXCLUSIVE subscriber only content.

George Crump is the Chief Marketing Officer at VergeIO, the leader in Ultraconverged Infrastructure. Prior to VergeIO he was Chief Product Strategist at StorONE. Before assuming roles with innovative technology vendors, George spent almost 14 years as the founder and lead analyst at Storage Switzerland. In his spare time, he continues to write blogs on Storage Switzerland to educate IT professionals on all aspects of data center storage. He is the primary contributor to Storage Switzerland and is a heavily sought-after public speaker. With over 30 years of experience designing storage solutions for data centers across the US, he has seen the birth of such technologies as RAID, NAS, SAN, Virtualization, Cloud, and Enterprise Flash. Before founding Storage Switzerland, he was CTO at one of the nation's largest storage integrators, where he was in charge of technology testing, integration, and product selection.

Tagged with: , , , , , , , , , ,
Posted in Briefing Note

Enter your email address to follow this blog and receive notifications of new posts by email.

Join 25,553 other subscribers
Blog Stats
%d bloggers like this: