Skip to content

PernixData

PernixData released FVP 3.5 and Architect 1.1

PernixData has released the latest version of their FVP and Architect software, bringing FVP to version 3.5 and Architect to version 1.1. Together with some small feature enhancements and support for vSphere 6.0 Update 2, the main enhancement is the virtual appliance.

pernixdata-virtual-appliance

I've deployed the appliance in my Homelab to test drive the appliance and their new features.

Read More »PernixData released FVP 3.5 and Architect 1.1

Evaluate PernixData FVP with replayed Production IO Traces

Using synthetic workloads to test drive PernixData FVP might result into odd findings. The most meaningful approach to test FVP is to deploy the software to production in monitor mode, let Architect do its magic and enable acceleration after checking the recommendations after a couple of days. Despite it is possible to deploy FVP, test drive, and remove it, without any downtime to virtual machines, this approach might not fit to all environments.

pernixdata-fvp-replay-workload

If you have separate DEV/QA environments with sophisticated load generators, the solution is obviously. If you don't have that, there is another option by record production I/O traces and replay them in a FVP accelerated test platform.

Read More »Evaluate PernixData FVP with replayed Production IO Traces

Howto NOT Benchmark PernixData FVP

Whenever I evaluate a new storage hardware or technology, I do some basic performance testing with VMware I/O Analyzer. I/O Analyzer is virtual appliance (Fling) provided by VMware that runs Iometer to generate synthetic I/O workloads. After installing PernixData FVP my first idea was to compare the raw performance of my storage, against the performance with PernixData FVP. However, I quickly noticed that synthetic workloads do not create any useful results. The upside is that I've learned much about how their caching operates so I'm publishing my results anyway. Actually, it's not a problem with PernixData FVP itself, it's just how host based caching works.pernixdata-fvp-vmware-io-analyzer

Conclusion: Don't use synthetic workloads to test caching solutions! PernixData FVP works as expected but synthetic I/O workloads are no meaningful benchmarks for host based caching. I am going to test with realworld applications in another article to show how they can take advantage of FVP.

Read More »Howto NOT Benchmark PernixData FVP

Getting Started with PernixData FVP Freedom

pernixdata-fvpPernixData offers a free version of their FVP Storage Accelerator. The Pernix Freedom licence is available for free and a good entry point to their technology. The Pernix Freedom license can be upgraded to a full version of FVP later.

Pernix Freedom Limitations

  • One FVP cluster only
  • RAM acceleration only
  • No SSD or PCIe acceleration
  • Read Acceleration Only (write-through)
  • 10 Minutes Performance Data
  • Community support only

Read More »Getting Started with PernixData FVP Freedom

PernixPro

Beside VMwares vExpert program where I am part of since 4 years I am proud to announce that I am now also part of the PernixPro program by PernixData.

florian-grehl-pernixpro

PernixPro™ members are an elite community of IT professionals committed to advancing FVP storage acceleration technology. By facilitating collaboration between PernixData experts and the industry’s top luminaries, PernixPro breaks down the barriers to widespread adoption of server-side storage, giving companies of all sizes a cost-effective solution for scaling storage performance independent of storage capacity.

PernixDatas main product, PernixData FVP, is a pure software based solution that creates a low latency, fault tolerant I/O acceleration tier using high-speed server-side Flash and/or Memory.

Read More »PernixPro