Today marks the release of the next evolution to the VxBlock, the VxBlock 1000. With this release, comes a departure from a few of the norms. Before I dive into the architectural changes, I want to highlight one of the biggest changes on the operations side. As of this announcement, the VxBlock 1000 is Ready to Ship (RTS). This means you can order it today. This is a change to the processes & procedures of old and one of the benefits of the new business model of DellEMC. Going forward, I would expect to see same day availability during future announcements as well.
So what is it? Well as I hinted to it in the title, this is a change from the previous VxBlock designs. Last week, there were 3 you could choose from (350, 540, & 740) and each came with a storage option, the 350 came with rack mount servers, and you could get various data protection options through tech extensions. Going forward, there will be only 1, the VxBlock 1000. With this new architecture you can combine various storage and server options to get exactly what you need. Compute options include both blade and rack mount cisco M4 & M5 servers. Storage options include the new Unity X50F, VMAX X50F, XtremIO X2 (coming later this year), and Isilon Gen 6. Combine in Avamar, Data Domain, and the DPS Suite as fully baked in supported options (which means tech extensions are no longer required). All of this is now covered under a single RCM as well. Changes also come to the management appliance (known as the new AMP-VX). It’s now smaller, cheaper, based on VSAN, and comes with integrated data protection.
Think of this as a simplification. No longer do you need to design a VxBlock for one or two workloads, you can now consolidate multiple solutions into a single system, allowing it all to be managed and supported together. Even if you need to have multiple VxBlock 1000s (which already scale to 14 cabinets each), a single AMP-VX cluster can manage up to 8 systems in total.
Personally, as a vArchitect for New England, I’ve been looking forward to this since I first heard about it internally. I think the simplification of the design means I can get exactly the right solution for my customer’s needs.
One year ago, EMC (yes it was still EMC at the time) launched their update to the VSPEX BLUE, EMC’s take on the EVO: Rail product. For those who don’t remember, EVO: Rail was a joint effort between VMware and several hardware vendors to release hyper converged infrastructure to the masses. I covered the original launch here and EMC’s spin on it. The EVO: Rail program was a success / failure depending on who you spoke to in the industry, but it did validate that HCI is here to stay and not just a passing fad. Later on, the EVO program was retooled under a joint venture between EMC & VMware under the VCE umbrella , and last year launched the evolution of that product … VxRail!
What went well?
VxRail went GA in March of 2016 and so far sales have good, especially in the mid market, especially on purpose built deployments (VDI, Databases, etc…). Working for a partner, I can attest that HCI adoption is going strong in the market place as more and more people look to it as a possible future. The product has undergone several major releases, which is huge for a product that was just starting to sell. It means that they are really listening to feedback and delivering to enhance the product regularly. It helps that VMware has been continuously driving innovation on the VSAN side (which VxRail is built on).
What needs improvement?
Dell EMC will be the first to admit that they didn’t expect the sales of this product (especially given the sales of the previous generation), and thus they didn’t quite bulk up the PS side to match. Unfortunately that mean delays while the back log was churned through. The other side was features was prioritized over serviceability, so things like the support gateway were not built in when you would have expected it was already included. The good news is things like this have been addressed (or are being addressed soon).
So what’s next for VxRail?
Flash Flash Flash! Recent sales have been leaning very heavily to all flash solutions, and that will be the trend going forward with VxRail sales which will also drive development on those lines especially in NVMe and NVDIMMs. The shift in the recent version to use Dell PowerEdge Servers has also opened up a lot of configuration options for memory and CPU.
With offering several different types of nodes, you can right size the product for the solution and make the price a lot more attractive (this is one of the things learned from the EVO sales).
Its great to see things are finally shaping up for the EVO program and just further validates something I’ve believed for years. Speaking as a partner, I welcome the extra options in HCI market space because it gives the customer a better chance to get what they want, and that will greatly help more people consider HCI an option for their next step in the data center journey. It was great to speak with Chad Dunn at Tech Field Day 13 (#TFD13) earlier this month in Austin. If you haven’t already, be sure to check out the videos from Dell EMC.