{"id":593,"date":"2012-05-08T16:11:28","date_gmt":"2012-05-08T22:11:28","guid":{"rendered":"http:\/\/benincosa.com\/blog\/?p=593"},"modified":"2014-11-19T11:24:53","modified_gmt":"2014-11-19T17:24:53","slug":"ucs-rainbows-6500s-and-the-soul-of-a-server","status":"publish","type":"post","link":"https:\/\/benincosa.com\/?p=593","title":{"rendered":"UCS: Rainbows, 6500s, and the Soul of a Server"},"content":{"rendered":"<p>My day job is to be an advocate for <a href=\"http:\/\/www.cisco.com\/en\/US\/products\/ps10265\/index.html\">Cisco UCS<\/a> in my customer&#8217;s data center. \u00a0Its a great gig. \u00a0Its much easier to back a product when you actually believe in it. \u00a0I thought I&#8217;d write down some of the ideas that I talk about with my customers on this blog.<\/p>\n<h2>Rainbows In the Data Center<\/h2>\n<p>Rack mount servers are still all the rage in many organizations. \u00a0And so is Gigabit Ethernet. \u00a0VMware <a href=\"http:\/\/www.vmware.com\/pdf\/Perf_Best_Practices_vSphere5.0.pdf\">best practices<\/a> suggest that you have separate networks for management, vMotion, I\/O, and VM traffic. \u00a0Using NIC Teaming you get something that looks like this beautiful picture:<\/p>\n<p>&nbsp;<\/p>\n<p><a href=\"http:\/\/benincosa.com\/blog\/wp-content\/uploads\/2012\/05\/RainbowsInDataCenter1.png\"><img decoding=\"async\" loading=\"lazy\" class=\"aligncenter size-full wp-image-595\" title=\"RainbowsInDataCenter\" src=\"http:\/\/benincosa.com\/blog\/wp-content\/uploads\/2012\/05\/RainbowsInDataCenter1.png\" alt=\"\" width=\"512\" height=\"366\" \/><\/a><\/p>\n<p style=\"text-align: center;\">(source: \u00a0Not sure, some Cisco person&#8217;s power point I stole)<\/p>\n<p style=\"text-align: left;\">You&#8217;ll notice that people color code these cables so they can tell which network goes to what. \u00a0The result is a beautiful rainbow flowing out of each server. \u00a0Then, <a href=\"http:\/\/en.wikipedia.org\/wiki\/Rainbow_Brite\">Rainbow Brite<\/a> and her sprite buddy Wink and her stallion Starlite can aid you in managing this big mess.<\/p>\n<p style=\"text-align: left;\">&nbsp;<\/p>\n<p><a href=\"http:\/\/benincosa.com\/blog\/wp-content\/uploads\/2012\/05\/RainbowBrite.png\"><img decoding=\"async\" loading=\"lazy\" class=\"aligncenter size-full wp-image-596\" title=\"RainbowBrite\" src=\"http:\/\/benincosa.com\/blog\/wp-content\/uploads\/2012\/05\/RainbowBrite.png\" alt=\"\" width=\"231\" height=\"316\" \/><\/a><\/p>\n<p>The account team at Cisco responsible for selling you switches loves this because for every cable you buy, you need to connect it to a switch port. \u00a0This is why we tell everybody that going with UCS is a strategic decision. \u00a0Do you want to continue investing in lots of Gigabit Ethernet Switches or consolidate with 10GbE? \u00a0UCS gets rid of the rainbows. \u00a0Yes, rainbows are pretty, but you don&#8217;t want them to get out of hand. \u00a0<a href=\"http:\/\/www.youtube.com\/watch?v=OQSNhk5ICTI\">Rainbows do strange things to people (especially if you have more than one)<\/a>. \u00a0Sometimes they can be just too much, and too intense.<\/p>\n<p>UCS gets you instant 10GbE as well as consolidation. \u00a0I did a comparison for a customer who was looking to buy a &#8220;pod&#8221; of rack mount servers and compared it to what it would take to buy the\u00a0equivalent\u00a0UCS. \u00a0Each &#8220;pod&#8221; consisted of 44 2-socket CPU servers. \u00a0Each server had 6 Gigabit Ethernet ports required as well as 2 HBAs to connect to the SAN. \u00a0The comparison was pretty eye opening. \u00a0By going with a UCS strategy the following benefits were realized:<\/p>\n<ul>\n<li>10% reduction in acquisition cost (some of this had to do with not having to buy new network switches)<\/li>\n<li>57% reduction in physical rack space<\/li>\n<li>A dramatic difference in Ethernet cables required: \u00a028 compared to 308<\/li>\n<li>A dramatic difference in Fibre Channel ports required: \u00a04 compared to 88<\/li>\n<\/ul>\n<p>Think this can happen with Legacy Non-UCS blades? \u00a0Not so much. \u00a0There is a savings still, but nothing as dramatic. \u00a0So if you like dealing with more infrastructure, stringing cables, and configuring port policies on your network switches for every server in your environment, UCS may not be for you.<\/p>\n<h2>UCS is the new Catalyst 6000<\/h2>\n<p>I get the chance to walk into the belly of many data centers. \u00a0One common feature that we see there is the venerable Cisco <a href=\"http:\/\/en.wikipedia.org\/wiki\/Catalyst_6500\">Catalyst 6000 switch. <\/a> Cisco has been milking this baby since 1999.<\/p>\n<p><a href=\"http:\/\/benincosa.com\/blog\/wp-content\/uploads\/2012\/05\/Venerable_65091.jpg\"><img decoding=\"async\" loading=\"lazy\" class=\"aligncenter size-full wp-image-598\" title=\"Venerable_6509\" src=\"http:\/\/benincosa.com\/blog\/wp-content\/uploads\/2012\/05\/Venerable_65091.jpg\" alt=\"\" width=\"220\" height=\"165\" \/><\/a><\/p>\n<p>What makes this thing so successful? \u00a0Removable line cards and supervisor line cards. \u00a0As people have migrated from fast ethernet to gigabit Ethernet to 10 Gigabit Ethernet they&#8217;ve just upgraded the line cards or supervisor modules. \u00a0They made the strategic decision years ago to go with this and its working out great.<\/p>\n<p>UCS Fabric Interconnects have a similar value proposition in the compute space. \u00a0You buy them as part of your strategy and then adding blades is just like adding remote line cards instead of fixed line cards in a 6509. \u00a0Going with this strategy provides several benefits:<\/p>\n<ul>\n<li>Cost effective way of getting servers on line. \u00a0These Cisco servers are extremely price competitive and very attractive. \u00a0Once you have the infrastructure, adding blades is so cost compelling its hard to see the rational of not going with another blade when you have the chance. \u00a0This isn&#8217;t to say you&#8217;re necessarily locked into Cisco. \u00a0You still have options. \u00a0Its just that the other options are not as attractive any more.<\/li>\n<li>Let&#8217;s suppose that 5 years from now Cisco decides it wants to start selling some esoteric micro servers in a new chassis or something. \u00a0(Full disclosure: \u00a0I have no idea if they are planning this and have seen nothing on the roadmap). \u00a0Let&#8217;s suppose that this new chassis has 100 slots for these servers. \u00a0If you bought any other blade, you&#8217;d need to throw out the old architecture and buy the new chassis. \u00a0With UCS, you just buy the new chassis and add these fun esoteric servers. \u00a0The Fabric Interconnects and the Fabric Extenders on the back will still work the same way. \u00a0 In essence: \u00a0You&#8217;ve future proofed your architecture.<\/li>\n<\/ul>\n<p>So what if everything goes 100GbE? \u00a0Fine, swap out Fabric Interconnects just like you swap out supervisor line cards on the 6500. \u00a0The architecture is brilliant. \u00a0What do you do with competitor solutions? \u00a0 You have to throw out, start over. \u00a0The Fabric Interconnect architecture is just something that builds and has the ability to build as time goes on.<\/p>\n<h2>The Soul of a Server<\/h2>\n<p>In addition to the nice architecture of UCS, another compelling feature is how we manage UCS servers. \u00a0The idea is a bit different than how you did things in the past. \u00a0Back in the day, when you wanted to set up a new server, you would plug it in, hook it up to a crash cart, turn it on and then press F2 \u00a0and do cool things like:<\/p>\n<ul>\n<li>Tune BIOS settings<\/li>\n<li>Set boot order<\/li>\n<li>Program iSCSI interfaces<\/li>\n<li>configure RAID<\/li>\n<\/ul>\n<p>Then you might go through and do several updates. \u00a0Of course you wrote it all down right? \u00a0Its not hard for these things to get out of sync. \u00a0And guess what causes problems in application performance? \u00a0When the infrastructure you thought was homogeneous is not homogeneous. \u00a0This is a pain and takes a lot more time than people readily admit.<\/p>\n<p>With UCS we do it differently. \u00a0Gone are the days of pressing F2. \u00a0I&#8217;ve never pressed F2 while a machine is booting on UCS Blades. \u00a0Here&#8217;s the new way: \u00a0There is a place in UCS Manager where all your wildest dreams and fantasies can come true. \u00a0This is where we logically define what we want our servers to look like. \u00a0We go through and say to ourselves: \u00a0In my fantasy world, if I could have a server, I&#8217;d want it to PXE boot, then boot to hard drive. \u00a0I&#8217;d want its BIOS settings to have hyperthreading. \u00a0I&#8217;d also like its firmware level to be 2.0(2) and I&#8217;d like its RAID to be set to RAID1 mirroring.<\/p>\n<p>That&#8217;s exactly how you do it. \u00a0You define a template that has all the characteristics of the server you want and then spawn instances of it (called Service Profiles). \u00a0You then take those spawns, or service profiles and possess the hardware of the physical blade you assign it to. \u00a0Its like you create the soul of the server then give that soul a body.<\/p>\n<p>This is pretty cool because now if you want to change something, you change it at the template and it can in turn update all the spawns of it. \u00a0You can create multiple templates for different types of servers, each optimized for the application that the server is supporting. \u00a0So you might have a service profile template for ESXi, a template for Oracle, a template for Windows bare metal, or a template for RHEV. \u00a0You are in the business of managing server souls. \u00a0Its more nobel.<\/p>\n<p>These are just a few of the many benefits of UCS that I thought I&#8217;d write down. \u00a0There&#8217;s always situations where other products may be more applicable. \u00a0But UCS is definitely one to check out. \u00a0And just in case you missed it: \u00a0UCS is the 3rd best selling x86 blade server world wide after HP (1) and IBM (2). \u00a0In the US, UCS is ranked #2 behind HP. \u00a0Not bad for a server that&#8217;s only been on the scene since 2009.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>My day job is to be an advocate for Cisco UCS in my customer&#8217;s data center. \u00a0Its a great gig. \u00a0Its much easier to back a product when you actually believe in it. \u00a0I thought I&#8217;d write down some of the ideas that I talk about with my customers on this blog. Rainbows In the&#8230;<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":[],"categories":[132,992],"tags":[],"jetpack_featured_media_url":"","_links":{"self":[{"href":"https:\/\/benincosa.com\/index.php?rest_route=\/wp\/v2\/posts\/593"}],"collection":[{"href":"https:\/\/benincosa.com\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/benincosa.com\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/benincosa.com\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/benincosa.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=593"}],"version-history":[{"count":3,"href":"https:\/\/benincosa.com\/index.php?rest_route=\/wp\/v2\/posts\/593\/revisions"}],"predecessor-version":[{"id":2773,"href":"https:\/\/benincosa.com\/index.php?rest_route=\/wp\/v2\/posts\/593\/revisions\/2773"}],"wp:attachment":[{"href":"https:\/\/benincosa.com\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=593"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/benincosa.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=593"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/benincosa.com\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=593"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}