<?xml version="1.0" encoding="utf-8"?>
<rss xmlns:atom="http://www.w3.org/2005/Atom" version="2.0"><channel><title>Gregs Server and StorageIOblog - Latest Comments</title><link xmlns="http://www.w3.org/2005/Atom" rel="http://api.friendfeed.com/2008/03#sup" href="http://disqus.com/sup/all.sup#forumcomments-2d8c25de" type="application/json"/><link>http://storageioblog.disqus.com/</link><description>Gregs Server and Storage IO Blog, enabling efficient and effective data infrastructures</description><atom:link href="http://storageioblog.disqus.com/comments.rss" rel="self"></atom:link><language>en</language><lastBuildDate>Wed, 22 May 2013 13:05:21 -0000</lastBuildDate><item><title>Re: EMC ViPR virtual physical object  and software defined storage (SDS)</title><link>http://storageioblog.com/emc-vipr-virtual-physical-object-software-defined-storage-sds/#comment-905028694</link><description>&lt;p&gt;Thanks and concur Loic, lets see how ViPR gets rolled out, who is included or has access and so forth...&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">greg schulz</dc:creator><pubDate>Wed, 22 May 2013 13:05:21 -0000</pubDate></item><item><title>Re: Part II: How many IOPS can a HDD, HHDD or SSD do with VMware?</title><link>http://storageioblog.com/part-ii-iops-hdd-hhdd-ssd/#comment-903306083</link><description>&lt;p&gt;FYI: Here is a link to a simple summary comparison of four HDDs, two are 2.5" 7,200 RPM (7.2K) desktop class (Seagate Momentus and a Fujitsu) and two 3.5" 7.2K drives (Seagate Barracuda and WD / Dell). The comparison simply shows at 4K IOP, single thread/worker, raw IO (this is a synopsis of those results shown above) how IOPs can vary across different physical size drives (granted these were different capacities) at the same speed.&lt;/p&gt;

&lt;p&gt;&lt;a href="http://storageio.com/DownloadItems/DevicePerformance/SIO_Perf_HDD_4KIOP_DriveSummary.pdf" rel="nofollow"&gt;http://storageio.com/DownloadI...&lt;/a&gt;&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">greg schulz</dc:creator><pubDate>Mon, 20 May 2013 16:29:39 -0000</pubDate></item><item><title>Re: EMC ViPR virtual physical object  and software defined storage (SDS)</title><link>http://storageioblog.com/emc-vipr-virtual-physical-object-software-defined-storage-sds/#comment-897900671</link><description>&lt;p&gt;Very good summary of the storage virtualisation that is now deployed for years mostly in upper segment of the market and now also available for the midrange and even in entry market.&lt;br&gt;Would be interesting to know what evolution ViPR will bring.&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Loic Boireau-Piazza</dc:creator><pubDate>Wed, 15 May 2013 04:00:25 -0000</pubDate></item><item><title>Re: EMC ViPR software defined object storage part III</title><link>http://storageioblog.com/emc-vipr-software-defined-object-storage-part-iii/#comment-894827542</link><description>&lt;p&gt;Greg, great informative article about VipR and very glad to see your mention about Zadara.&lt;/p&gt;

&lt;p&gt;Kudos to EMC on the thought leadership, I know how hard is for the incumbent to innovate on his own space.&lt;/p&gt;

&lt;p&gt;At Zadara Storage, we had the big advantage of starting from scratch with a new storage architecture that allows us to bring some of the most important capabilities of SDS.&lt;/p&gt;

&lt;p&gt;1 - multi tenancy without compromises.&lt;br&gt;Zadara allows separation of resources and management between the different tenants in a multitenant environment. (each tenant feels they run alone with full control and performance of their virtual storage array)&lt;/p&gt;

&lt;p&gt;2 - pure software run on standard x86 hardware&lt;br&gt;With Zadara, not only the control path but also the data path use standard x86 hardware, we achieve similar performance and better reliability than custom SAN/NAS storage arrays. &lt;/p&gt;

&lt;p&gt;3 - security - Zadara's Virtual storage arrays are completely separated one to each other by different VLANs. (VLANs and IP addresses are automatically allocated using SDN capabilities)&lt;/p&gt;

&lt;p&gt;4 - scale ability - Scale from just few to thousands of VIrtual Storage Arrays running concurrently on the same hardware with rolling upgrades capabilities of the entire system.&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Nelson Nahum</dc:creator><pubDate>Sun, 12 May 2013 13:59:48 -0000</pubDate></item><item><title>Re: Cloud conversations: AWS EBS, Glacier and S3 overview (Part I)</title><link>http://storageioblog.com/cloud-conversations-aws-ebs-glacier-and-s3-overview-part-i/#comment-890364059</link><description>&lt;p&gt;An update, today (5/8/13) AWS announced that Provisioned IOPS have been raised from 2,000 per EBS volume to 4,000 per EBS volume. Note also that with AWS provisioned IOPS that IO size great than 16KB count as two IOPs, for example an 8KB IOP is one AWS IOP, a 32KB IOP is 2 IOPS, a 64GB IOP is 4 AWS IOPS etc. Likewise latency or response time can be variable vs. deterministic.&lt;/p&gt;

&lt;p&gt;Also note that if you need more performance than by stripe multiple EBS volumes together, another option is to use the EC2 high IO instance that has 2 x 100GB+ SSDs.&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">greg schulz</dc:creator><pubDate>Thu, 09 May 2013 00:15:03 -0000</pubDate></item><item><title>Re: EMC ViPR software defined object storage part III</title><link>http://storageioblog.com/emc-vipr-software-defined-object-storage-part-iii/#comment-889119492</link><description>&lt;p&gt;Great, thanks Chad, had some more good NDA convos today here at EMCworld with some of your associates from ASD, will drop you a line if need anything else. - gs&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">greg schulz</dc:creator><pubDate>Wed, 08 May 2013 01:33:09 -0000</pubDate></item><item><title>Re: EMC ViPR software defined object storage part III</title><link>http://storageioblog.com/emc-vipr-software-defined-object-storage-part-iii/#comment-889101286</link><description>&lt;p&gt;Disclosure - EMCer here.&lt;/p&gt;

&lt;p&gt;Greg - a very good blog series (and fair and balanced, including some of your questions).   Happy to answer any Qs I can - drop me a line!&lt;/p&gt;

&lt;p&gt;Chad&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Chad Sakac</dc:creator><pubDate>Wed, 08 May 2013 01:24:17 -0000</pubDate></item><item><title>Re: Little data, big data and very big data (VBD) or big BS?</title><link>http://storageioblog.com/little-data-big-data-and-very-big-data-vbd-or-big-bs/#comment-877920798</link><description>&lt;p&gt;Nice work Greg - you did it again!&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Marc Farley</dc:creator><pubDate>Sat, 27 Apr 2013 11:58:50 -0000</pubDate></item><item><title>Re: Where has the FCoE hype and FUD gone? (with poll)</title><link>http://storageioblog.com/where-has-the-fcoe-hype-and-fud-gone-with-poll/#comment-877146046</link><description>&lt;p&gt;There is an interesting theme here in that old info can be come modern fud. For example there are (believe it or not) those that are surprised to hear Ethernet or what they may know as 802.x can run on Fiber and is no longer the collision detect based model back form 10/100 era. Likewise while much of Ethernet is more plug and play than in the past, there are also different gradients or grades of technology and capabilities that go with them.&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">greg schulz</dc:creator><pubDate>Fri, 26 Apr 2013 16:51:21 -0000</pubDate></item><item><title>Re: Where has the FCoE hype and FUD gone? (with poll)</title><link>http://storageioblog.com/where-has-the-fcoe-hype-and-fud-gone-with-poll/#comment-877144039</link><description>&lt;p&gt;@StorageTechie concur that some people are slower to adopt for various reasons, some are risk averse (or have to be), some are just cautious, some skeptical. Otoh there are those who will race to something new, sometimes dropping what they were doing or using just to get their hands on the shiny new tech toy. Concur that over time (years not months) more of the current FC usage will shift over to FCoE with some of that also going to iSCSI or NAS.&lt;/p&gt;

&lt;p&gt;Speaking of NAS, there are also those who look at block heads (e.g. iSCSI, SAS, FC, FCoE, IBA/SRP, etc) as being backwards or in the past, just as there are those who see anybody using block of nas/file (NFS, CIFS, AFP, AFS, HDFS, etc) as being backwards vs. using object APIs and vise versa.&lt;/p&gt;

&lt;p&gt;Otoh, its nice to have options to use for various things...&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">greg schulz</dc:creator><pubDate>Fri, 26 Apr 2013 16:48:29 -0000</pubDate></item><item><title>Re: Where has the FCoE hype and FUD gone? (with poll)</title><link>http://storageioblog.com/where-has-the-fcoe-hype-and-fud-gone-with-poll/#comment-875572099</link><description>&lt;p&gt;Untrue. Ethernet can be very reliable. Just depends 1st on if you're using the right equipment for the right spot in the network. I.e. a desktop switch for your top of rack is NOT right; next does it have the appropriate features - queues, buffers, chips; finally - did you implement it right, or just throw it in and assume that since you provided a lot of bandwidth that you'll be fine. The old way of implementing Ethernet (throwing it into place and turning it on) is not OK anymore. It actually requires a little thought and planning. But when you do it right - it's damned reliable.&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">sheep1451</dc:creator><pubDate>Thu, 25 Apr 2013 13:24:35 -0000</pubDate></item><item><title>Re: Where has the FCoE hype and FUD gone? (with poll)</title><link>http://storageioblog.com/where-has-the-fcoe-hype-and-fud-gone-with-poll/#comment-875553330</link><description>&lt;p&gt;I think that as people start to see the efficiency differences between even 10G FCoE vs. 16G FC, things will start to move increasingly away from FC and towards FCoE. When you start to look at what 40G FCoE can do vs. something like 32G FC, it's a no-brainer. But there will ALWAYS be people that will be slow to adopt it b/c they simply don't need to - what they have is good enough. Or they're a few years from retirement. Or... Or... As the speeds go up, and the cost of FC climbs with it, you'll start seeing more movement to FCoE. I predict anyway.&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">sheep1451</dc:creator><pubDate>Thu, 25 Apr 2013 13:03:20 -0000</pubDate></item><item><title>Re: Part II: XtremIO,  XtremSW and XtremSF EMC flash ssd portfolio redefined</title><link>http://storageioblog.com/part-ii-xtremio-xtremsw-and-xtremsf-emc-flash-ssd-portfolio-redefined/#comment-873157711</link><description>&lt;p&gt;Good and details information. SSD xtem0 and xteamsf flash are realible services.&lt;/p&gt;

&lt;p&gt;&lt;a href="http://www.cwps.com/" rel="nofollow"&gt;http://www.cwps.com/&lt;/a&gt;&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">James Taylor</dc:creator><pubDate>Tue, 23 Apr 2013 07:58:35 -0000</pubDate></item><item><title>Re: How a pressure cooker should be used for good things</title><link>http://storageioblog.com/?p=4829#comment-871821443</link><description>&lt;p&gt;Thks Louis, hope all is well...&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">greg schulz</dc:creator><pubDate>Mon, 22 Apr 2013 00:32:23 -0000</pubDate></item><item><title>Re: How a pressure cooker should be used for good things</title><link>http://storageioblog.com/?p=4829#comment-871816977</link><description>&lt;p&gt;You are out of your mind, Greg. Good stuff, sir.&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Louis Gray</dc:creator><pubDate>Mon, 22 Apr 2013 00:26:19 -0000</pubDate></item><item><title>Re: Little data, big data and very big data (VBD) or big BS?</title><link>http://storageioblog.com/little-data-big-data-and-very-big-data-vbd-or-big-bs/#comment-865253194</link><description>&lt;p&gt;Thanks Rebecca, perhaps Big Data is in the eyes of the beholder?&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">greg schulz</dc:creator><pubDate>Tue, 16 Apr 2013 12:09:50 -0000</pubDate></item><item><title>Re: Little data, big data and very big data (VBD) or big BS?</title><link>http://storageioblog.com/little-data-big-data-and-very-big-data-vbd-or-big-bs/#comment-865244283</link><description>&lt;p&gt;Great post, Greg.  I eagerly await your industry trends and perspectives on Software Defined Whatever (SDx).  Never a dull moment in marketing . . .&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Rob Commins</dc:creator><pubDate>Tue, 16 Apr 2013 11:59:15 -0000</pubDate></item><item><title>Re: HP Moonshot 1500 software defined capable compute servers</title><link>http://storageioblog.com/hp-moonshot-1500-software-defined-capable-compute-servers/#comment-860666086</link><description>&lt;p&gt;Software-defined can be interpreted in different ways. HP is defining servers for specific applications (software), hence the wording software-defined servers. Of course its also convenient from a marketing perspective to use buzz words but in this case, the wording is valid but has a different reasoning behind it.&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">easyisez</dc:creator><pubDate>Fri, 12 Apr 2013 05:10:16 -0000</pubDate></item><item><title>Re: Where has the FCoE hype and FUD gone? (with poll)</title><link>http://storageioblog.com/where-has-the-fcoe-hype-and-fud-gone-with-poll/#comment-860063891</link><description>&lt;p&gt;Here is a link to a copy of this post over at the VMware communities site along with some comments/discussion.&lt;/p&gt;

&lt;p&gt;&lt;a href="http://communities.vmware.com/people/gregschulz/blog/2013/04/01/where-has-the-fcoe-hype-and-fud-gone-with-poll#comment-28381" rel="nofollow"&gt;http://communities.vmware.com/...&lt;/a&gt;&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">greg schulz</dc:creator><pubDate>Thu, 11 Apr 2013 17:18:45 -0000</pubDate></item><item><title>Re: HP Moonshot 1500 software defined capable compute servers</title><link>http://storageioblog.com/hp-moonshot-1500-software-defined-capable-compute-servers/#comment-857228545</link><description>&lt;p&gt;:%s/Nicera/Nicira&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">larstr</dc:creator><pubDate>Tue, 09 Apr 2013 02:37:54 -0000</pubDate></item><item><title>Re: SNIA Spring 2013 update with Wayne Adams</title><link>http://storageioblog.com/snia-spring-2013-update-with-wayne-adams/#comment-853679167</link><description>&lt;p&gt;Hello, i was wondering if anyone here knows of a good server company in   ottawa i just need &lt;a href="http://www.boyd.ca/s_storage.html" rel="nofollow"&gt;storage&lt;/a&gt; space, and i might run a game off of it as well Thanks!&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Jason Knight</dc:creator><pubDate>Fri, 05 Apr 2013 21:51:14 -0000</pubDate></item><item><title>Re: Where has the FCoE hype and FUD gone? (with poll)</title><link>http://storageioblog.com/where-has-the-fcoe-hype-and-fud-gone-with-poll/#comment-853458293</link><description>&lt;p&gt;@StorageTechie thanks for the comments and perspectives. Do you think that there will be a time where some non "frugal" shops look at 10GbE or 40GbE for moving up to higher performance, or vise versa, higher-end environments becoming more "frugal" looking at iSCSI either for actual adoption or to gain leverage with FC vendors?&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">greg schulz</dc:creator><pubDate>Fri, 05 Apr 2013 16:19:20 -0000</pubDate></item><item><title>Re: Where has the FCoE hype and FUD gone? (with poll)</title><link>http://storageioblog.com/where-has-the-fcoe-hype-and-fud-gone-with-poll/#comment-852249538</link><description>&lt;p&gt;I still see a BUNCH of Fibre Channel out there, and the 32-Gb standard will likely be finalized this year.   For shops big enough to bear the expense, I think FC is going to stick around as long as it meets performance demands.   For those shops that don't want to invest in FC, then iSCSI is a cheaper alternative - and has finally matured enough that people aren't afraid of it anymore.&lt;br&gt;So where's the room in the marketplace for FCoE?&lt;br&gt;The "frugal" shops have a cheaper alternative, while the shops willing spend more have a more mature alternative that (right now) is out-performing MOST other technologies.&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">StorageTechie</dc:creator><pubDate>Thu, 04 Apr 2013 16:57:24 -0000</pubDate></item><item><title>Re: Where has the FCoE hype and FUD gone? (with poll)</title><link>http://storageioblog.com/where-has-the-fcoe-hype-and-fud-gone-with-poll/#comment-845121127</link><description>&lt;p&gt;FCoE is as bad of an idea as iSCSI was and is...... Ethernet is the least reliable connection of all and people expect to run time sensitive scsi commands over it, what a sad joke.&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Guy Falkes</dc:creator><pubDate>Thu, 28 Mar 2013 11:43:01 -0000</pubDate></item><item><title>Re: EMC VFCache respinning SSD and intelligent caching (Part II)</title><link>http://storageioblog.com/emc-vfcache-respinning-ssd-and-intelligent-caching-part-ii/#comment-843903599</link><description>&lt;p&gt;Ray there are SSD cache and target cards for Cisco blades from others besides FIO have a look at these:&lt;br&gt;&lt;a href="http://www.emc.com/about/news/press/2012/20120828-02.htm" rel="nofollow"&gt;http://www.emc.com/about/news/...&lt;/a&gt;&lt;/p&gt;

&lt;p&gt;&lt;a href="http://www.lsi.com/about/newsroom/Pages/20120523pr.aspx" rel="nofollow"&gt;http://www.lsi.com/about/newsr...&lt;/a&gt;&lt;br&gt;Cheers gs&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">greg schulz</dc:creator><pubDate>Wed, 27 Mar 2013 09:44:50 -0000</pubDate></item></channel></rss>