Archive

Posts Tagged ‘Melio’

#Citrix #PVS vs. #MCS Revisited – #Nutanix, #Sanbolic

Another good blog post from Citrix and Nick Rintalan around the famous topic whether to go for PVS or MCS! If your thinking about this topic then don’t miss this article. Also ensure that you talk to someone who have implemented an image mgmt/provisioning service like this to get some details on lessons learnt etc., also with the change in the hypervisor layer and the cache features this is getting really interesting…

AND don’t forget the really nice storage solutions that exists out there like Nutanix and Melio that really solves some challenges out there!!

Nutanix, how it works..

http://go.nutanix.com/rs/nutanix/images/TG_XenDesktop_vSphere_on_Nutanix_RA.pdf

Melio Solutions – Virtual Desktop Infrastructure

Back to the Citrix blog post:

It’s been a few months since my last article, but rest assured, I’ve been keeping busy and I have a ton of stuff in my head that I’m committed to getting down on paper in the near future.  Why so busy?  Well, our Mobility products are keeping me busy for sure.  But I also spent the last month or so preparing for 2 different sessions at BriForum Chicago.  My colleague, Dan Allen, and I co-presented on the topics of IOPS and Folder Redirection.  Once Brian makes the videos and decks available online, I’ll be sure to point people to them.

So what stuff do I want to get down on paper and turn into a future article?  To name a few…MCS vs. PVS (revisited), NUMA and XA VM Sizing, XenMobile Lessons Learned “2.0″, and Virtualizing PVS Part 3.  But let’s talk about that first topic of PVS vs MCS now.

Although BriForum (and Synergy) are always busy times, I always try to catch a few sessions by some of my favorite presenters.  One of them is Jim Moyle and he actually inspired this article.  If you don’t know Jim, he is one of our CTPs and works for Atlantis Computing – he also wrote one of the most informative papers on IOPS I’ve ever read.  I swear there is not a month that goes by that I don’t get asked about PVS vs. MCS (pros and cons, what should I use, etc.).  I’m not going to get into the pros and cons or tell you what to use since many folks like Dan Feller have done a good job of that already, even with beautiful decision trees.  I might note that Barry Schiffer has an updated decision tree you might want to check out, too.  But I do want to talk about one of the main reasons people often cite for not using MCS – it generates about “1.6x or 60% more IOPS compared to PVS“.  And ever since Ken Bell sort of “documented” this in passing about 2-3 years ago, that’s sort of been Gospel and no one had challenged it.  But our CCS team was seeing slightly different results in the field and Jim Moyle also decided to challenge that statement. And Jim shared the results of his MCS vs. PVS testing at BriForum this year – I think many folks were shocked by the results.

What were those results?  Here is a summary of the things I thought were most interesting:

  • MCS generates 21.5% more average IOPS compared to PVS in the steady-state (not anywhere near 60%)
  • This breaks down to about 8% more write IO and 13% more read IO
  • MCS generates 45.2% more peak IOPS compared to PVS (this is closer to the 50-60% range that we originally documented)
  • The read-to-write (R/W) IO ratio for PVS was 90%+ writes in both the steady-state and peak(nothing new here)
  • The R/W ratio for MCS at peak was 47/53 (we’ve long said it’s about 50/50 for MCS, so nothing new here)
  • The R/W ratio for MCS in the steady-state was 17/83 (this was a bit of a surprise, much like the first bullet)

So how can this be?!?

I think it’s critical to understand where our initial “1.5-1.6x” or “50-60%” statement comes from – that takes into account not just the steady-state, but also the boot and logon phases, which are mostly read IOPS and absolutely drive up the numbers for MCS.  If you’re unfamiliar with the typical R/W ratios for a Windows VM during the various stages of its “life” (boot, logon, steady-state, idle, logoff, etc.), then this picture, courtesy of Project VRC, always does a good job explaining it succinctly:

The R/W ratio of the boot phase is a lot different than the steady-state!

 

We were also looking at peak IOPS and average IOPS in a single number – we didn’t provide two different numbers or break it down like Jim and I did above in the results, and a single IOPS number can be very misleading in itself.  You don’t believe me?  Just check out my BriForum presentation on IOPS and I’ll show you several examples of how…

Continue reading here!

//Richard

#Sanbolic Brings Public Cloud Economics to the Enterprise – #Melio

March 18, 2013 1 comment

Ok, I must say that this product is great!!! If you haven’t looked at it before then please do! And contact us at EnvokeIT if you want more details!

Sanbolic Enables Distributed Flash, SSD and HDD to Achieve Enterprise Systems Capability and Scale-Out In Server-Side and Commodity Storage Deployments

Waltham, MA – (March 18, 2013) – Sanbolic® today announced the general availability of its Melio version 5 (Melio5™) software – delivering distributed scale-out, high-availability and enterprise data services through software. Server-side flash has seen rapid adoption for applications such as hyperscale web serving, but limited adoption in general purpose enterprise applications. With the launch of Melio5, Sanbolic enables enterprise customers to dramatically improve their storage infrastructure economics by enabling server-side flash, SSD and HDD as primary persistent storage. Melio5 aggregates across nodes for scale-out and availability while providing RAID, remote replication, quality of Service (QoS), snapshots and systems functionality through a software layer on commodity hardware. This provides customers with the ability to deploy commodity and server-based storage architecture with similar economics and flexibility as public cloud data centers such as Google and Facebook.

With validation by hundreds of enterprise and government organizations running in production, Melio volume management and file system technology addresses the needs of high performing cost effective storage infrastructure on-premise. Melio5’s architecture is designed to scale up to 2,048 nodes and up to 65,000 storage devices enabling linear performance scalability in a cluster.

Melio5 also eliminates the need to deploy a redundant flash caching layer in front of legacy storage area network (SAN) hardware by directly incorporating flash into hybrid volumes and intelligently placing data based on file system access profiles. A hybrid volume will place random access data such as file system metadata on flash sectors while placing sequential data on low cost hard disk drives to greatly reduce the cost of capacity. The result is a highly scalable, high performance storage system, with a much lower cost than legacy storage arrays.

“Typically, server and disk drive vendors operate on gross margins in the 20-30% range. Storage array vendors, on the other hand, are often twice that or more,” said Eric Slack, Senior Analyst,Storage Switzerland. “Sanbolic’s approach leverages the architecture that the big social media and public cloud companies use, to fix this problem. By replacing storage arrays (and storage array margins) with commodity server and disk drive hardware and enabling it with intelligence through software, companies can significantly reduce storage infrastructure costs.”

Terri McClure, Senior Analyst, Enterprise Strategy Group (ESG), stated, “Sanbolic’s Melio5 software enables corporate users to take advantage of flash and SSD in conjunction with commodity hardware to create an intelligent, cost effective, and high performance storage architecture like the huge public cloud companies run, while still ensuring enterprise workload scalability and high availability.”

“Melio5 lets us solve one of the biggest challenges for our customers today – the upfront and management cost for storage – without sacrificing systems capability or performance. The Lego-like modular capability of Melio allows our customers to scale-out their storage and servers based on off-the-self commodity components, without downtime,” said Mattias Tornblom, CEO, EnvokeIT.

“LSI and Sanbolic’s shared vision and complementary products help customers to dramatically improve the performance, flexibility and economics of their on-premise storage infrastructure,” said Brent Blanchard, Senior Director of Worldwide Channel Sales and Marketing, LSI Corporation. “LSI’s Nytro™ family of server-side flash acceleration cards and leading SAS-based server storage connectivity solutions…

Continue reading here or here!

//Richard

My Post-Synergy View – Update 2 #CitrixSynergy #Citrix #CitrixSummit #XenApp #XenDesktop #NetScaler #Sanbolic

October 23, 2012 Leave a comment

Hi again!

Ok, time to wrap up my Synergy update post series… In the previous post I started to give U my view of the takeaways from Citrix Synergy. So let’s continue! 🙂

As I wrote before I tried to structure the post using the following; each heading represents the enhancement request topic and/or the takeaway item, and then the subheading of Description and Status is showing you my personal view on the topic and its status.

NetScaler

Description:

These where the items that we had on the enhancement list before going to Synergy:

  1. Create SDX platform/architecture to run on all MPX appliances, for more info why see; NetScaler MPX vs. SDX dilemma; https://richardegenas.com/2012/10/03/netscaler-mpx-vs-sdx-dilemma/
  2. Add support for AG session policies so that ICA proxy can be turned on for specific published apps and desktops and not per session. This for situations where you might have one app or desktop that sits behind an AGEE and others don’t.
  3. The NetScaler/Access Gateway HTML/GUI pages used shall be able to be customized per AGEE/AAA Virtual Server. Today they are global pages so that specific modifications/customizations cannot be made and you have to buy an additional NetScaler unless major customizations are done and then life-cycle management becomes an issue.
  4. Change so that you can specify different Authentication policies and requirements mapped to Session policies instead of to a Virtual Server, AAA group etc. This could then provide a way so that you could offer ICA proxy mode with single auth and two-factor if you launch/select to open an SSL VPN tunnel. And then a user that has forgotten a hardtoken could still get access but only in ICA proxy mode and have all virtual channels disabled without having to have multiple accounts in the Receiver and admin doesn’t need multiple NS AGEE VS.
  5. It would be good to get the same Account Self-Service functionality that the  Single Sign-On/Password Manager service can provide and have integrated into the AGEE login page where end-users can themselves unlock their accounts and reset their passwords without involving Service Desk. And the solution for how to get the single sign-on account self service feature should be an integration part of the NetScaler AGEE console rather than manually updating the HTML pages etc.

Read more…

%d bloggers like this: