On February 16th 2016, EMC has launched a new product: VxRail. It’s a fully integrated, preconfigured, and pre-tested VMware hyper-converged infrastructure appliancebased on VMware’s vSphere and Virtual SAN, and EMC software.

VxRail delivers an all-in-one IT infrastructure transformation by leveraging a known and proven building block for the Software Defined Data Center (SDDC).

But more important is not only an EMC product: it cames from the VCE that is an alliance with EMC, VMware and Cisco. So it’s mainly an EMC and VMware new product.

This announce brings some interesting questions related both to VMware SDDC strategy, but also with the new Dell deal.

First related to VMware, is what will happen to EVO:RAIL (and the entire EVO family)? EVO:RAIL is a great solution and really easy to setup, but this new appliance make the same things. Considering also how bad is performing EVO:RAIL on the marking and sales aspects, and how VMware has change the focus from the infrastructure to the services layer (in the last 12 months there were not really new news about EVO), seems that EVO project it’s dead. For sure does not have good time and unless something is changing on marketing and costs aspects will be difficult to see a good future for EVO family. At least we have really good reference design and architecture from the EVO:SDDC project, but considering that there is also a VxRack announce (a complete “block” from VCE based on VxRail), also EVO:SDDC does not have a clear future at all.

About the Dell aspect, it seems that those appliances are not based on Dell hardware and neither on Cisco UCS (EMC has build a specific server solution for the first version of it’s EVO:RAIL appliance, and this could be a similar approach). Also Dell has it’s own hyperconverged strategy with Nutanix parthership and will be really interesting see what will happen both for EMC and Dell products (and strategies) in the next months.

Going back to VxRail, there are two different type of nodes:

  • All-Flash Nodes – Maximum performance and scale for mission critical workloads and applications that demand low latency
  • Hybrid Nodes – Blends performance and cost across a range of processing power, memory and storage

The number of models are more depending also the size of memory, type of processors and raw space:

VxRail 60 VxRail 120 VxRail 160 VxRail 200
Processor Cores 
per node 
6 12 16 20
Processors 
per node 
1 Intel® Xeon® Processor 
ES-2603 v3 1.6 GHz
2 Intel® Xeon® Processor 
ES-2620 v3 2.4 GHz
2 Intel® Xeon® Processor 
ES-2630 v3 2.4 GHz
2 Intel® Xeon® Processor 
ES-2660 v3 2.6 GHz
RAM 
per node 
64 GB 
(4 x 16 GB) 
128 GB or 192 GB 
(8 x 16 GB) or (12 x 16 GB) 
256 GB or 512 GB 
(16 x 16 GB) or (16 x 32 GB)
256 GB or 512 GB 
(16 x 16 GB) or (16 x 32 GB)
Storage (Raw) 
per node 
3.6-6 TB 3.6-6 TB 4.8-6 TB 4.8-6 TB
Minimum Nodes Per Cluster 4 4 4 4
Maximum Nodes Per Cluster 64 64 64 64
Scaling Increments 
in nodes 
4 4 4 4
Andrea MauroAbout Andrea Mauro (2448 Posts)

Virtualization & Cloud Architect. VMUG IT Co-Founder and board member. VMware VMTN Moderator and vExpert (2010, 2011, 2012, 2013, 2014, 2015). PernixPro 2014. Dell TechCenter Rockstar 2014. MVP 2014. Several certifications including: VCDX-DCV, VCP-DCV/DT/Cloud, VCAP-DCA/DCD/CIA/CID/DTA/DTD, MCSA, MCSE, MCITP, CCA, NPP.


Share