{"id":9496,"date":"2016-06-06T12:27:01","date_gmt":"2016-06-06T19:27:01","guid":{"rendered":"https:\/\/www.aerocominc.com\/info\/?p=9496"},"modified":"2017-05-11T09:35:07","modified_gmt":"2017-05-11T16:35:07","slug":"cloud-therapy-ep-005-improving-your-cloud-server-implementation-with-raj-palat","status":"publish","type":"post","link":"https:\/\/www.aerocominc.com\/info\/cloud-therapy-ep-005-improving-your-cloud-server-implementation-with-raj-palat\/","title":{"rendered":"Cloud Therapy: EP 005 &#8211; Improving your Cloud Server Implementation with Raj Palat"},"content":{"rendered":"<p><a href=\"https:\/\/www.aerocominc.com\/info\/wp-content\/uploads\/2016\/06\/Cloud-Therapy-E005-LinkedIn.jpg\" rel='magnific'><img loading=\"lazy\" class=\"size-full wp-image-9752 aligncenter\" src=\"https:\/\/www.aerocominc.com\/info\/wp-content\/uploads\/2016\/06\/Cloud-Therapy-E005-LinkedIn.jpg\" alt=\"Cloud Therapy E005 - LinkedIn\" width=\"700\" height=\"400\" srcset=\"https:\/\/www.aerocominc.com\/info\/wp-content\/uploads\/2016\/06\/Cloud-Therapy-E005-LinkedIn.jpg 700w, https:\/\/www.aerocominc.com\/info\/wp-content\/uploads\/2016\/06\/Cloud-Therapy-E005-LinkedIn-300x171.jpg 300w, https:\/\/www.aerocominc.com\/info\/wp-content\/uploads\/2016\/06\/Cloud-Therapy-E005-LinkedIn-250x143.jpg 250w, https:\/\/www.aerocominc.com\/info\/wp-content\/uploads\/2016\/06\/Cloud-Therapy-E005-LinkedIn-600x343.jpg 600w\" sizes=\"(max-width: 700px) 100vw, 700px\" \/><\/a><\/p>\n<p>In this episode, <span style=\"color: #ff6600;\"><a style=\"color: #ff6600;\" href=\"http:\/\/www.aerocominc.com\/company-profile\/evolve-ip\" target=\"_blank\">Evolve IP<\/a><\/span>&#8216;s Manager of Sales Engineering, <a href=\"https:\/\/www.linkedin.com\/in\/rpalat\/\" target=\"_blank\">Raj Palat<\/a>, outlines his insightful &#8220;Data Center Diamond&#8221; approach to <span style=\"color: #ff6600;\"><a style=\"color: #ff6600;\" href=\"https:\/\/www.aerocominc.com\/info\/product\/cloud\/\" target=\"_blank\">cloud<\/a><\/span> implementation. It&#8217;s a holistic approach to evaluating your cloud decision from every angle, which is exactly what your boss will love to see.<\/p>\n<p>For more episodes, find us on <a href=\"https:\/\/itunes.apple.com\/us\/podcast\/cloud-therapy-aerocominc.com\/id1112772590?mt=2\" target=\"_blank\">iTunes<\/a> and <a href=\"http:\/\/www.stitcher.com\/podcast\/cloud-therapy-with-aerocominccom?refid=stpr\" target=\"_blank\">Stitcher<\/a>!<\/p>\n<p><iframe style=\"border: none;\" src=\"\/\/html5-player.libsyn.com\/embed\/episode\/id\/4397916\/height\/90\/width\/640\/theme\/custom\/autonext\/no\/thumbnail\/yes\/autoplay\/no\/preload\/no\/no_addthis\/no\/direction\/backward\/render-playlist\/no\/custom-color\/87A93A\/\" width=\"640\" height=\"90\" scrolling=\"no\" allowfullscreen=\"allowfullscreen\"><\/iframe><\/p>\n<p>CLICK BELOW\u00a0to learn which 3 cloud options are the best fit for your company.<\/p>\n<p><script src=\"\/\/static.leadpages.net\/leadboxes\/current\/embed.js\" async=\"\" defer=\"defer\"><\/script><button style=\"background: #f26e22; border-color: #f26e22; border-radius: 4px; color: #ffffff; display: inline-block; vertical-align: middle; padding: 16px 32px; min-width: 192px; border: 1px solid #f26e22; font-size: 1rem; font-family: Helvetica, Arial, sans-serif; text-align: center; outline: 0; line-height: 1; cursor: pointer; -webkit-transition: background 0.3s, color 0.3s, border 0.3s; transition: background 0.3s, color 0.3s, border 0.3s; box-shadow: 0px 5px 5px rgba(0, 0, 0, 0.6);\" data-leadbox-popup=\"144249073f72a2:136337e37f46dc\">MY TOP 3<\/button> \u00a0 <\/p>\n<h5><em>See full transcript below:<\/em><\/h5>\n<p>\u00a0 Mike: \u00a0\u00a0 <a href=\"https:\/\/aerocominc.lpages.co\/leadbox\/1419819f3f72a2%3A136337e37f46dc\/5704147139559424\/\" target=\"_blank\"><img class=\"alignright\" src=\"https:\/\/lh3.googleusercontent.com\/RB-xlCBQ9RJ3324Drdj1avFds6qFpI54rToFB1RpABQxRsBxoYdZZjmYtZ4jpb5ZvtOyI9Dgb0SVc5tv95OwTQ=s0\" alt=\"\" \/><\/a><script src=\"https:\/\/aerocominc.lpages.co\/leadbox-1493060101.js\" type=\"text\/javascript\" data-leadbox=\"1419819f3f72a2:136337e37f46dc\" data-url=\"https:\/\/aerocominc.lpages.co\/leadbox\/1419819f3f72a2%3A136337e37f46dc\/5704147139559424\/\" data-config=\"%7B%7D\"><\/script>So, one of your company\u2019s internal applications or software applications is running slow, and the higher-ups come to you and tell you it\u2019s time to upgrade the server and that they, possibly, want to look in to a cloud solution for that server. What if within 30 days, you could collect a ton of good information and present a report to your boss that details every aspect that he or she needs to know about that project? Do you think your boss would be impressed? Do you think that would go pretty far with him or her and earn you some brownie points for the future? You bet it would.<\/p>\n<p>How could you do that and not have it take up an entire six months of work? Well, today\u2019s podcast is going to tell you a little bit of how you might be able to put that together. Today, we have a special guest, his name is <a href=\"https:\/\/www.linkedin.com\/in\/rpalat\/\" target=\"_blank\">Raj Palat<\/a>. He\u2019s the manager of sales engineering for a great cloud company called <span style=\"color: #ff6600;\"><a style=\"color: #ff6600;\" href=\"http:\/\/www.aerocominc.com\/company-profile\/evolve-ip\" target=\"_blank\">Evolve IP<\/a><\/span>.<\/p>\n<p>Now, if you\u2019ve heard of Evolve IP, obviously, you know a little bit about him. If you\u2019ve not heard of him, they have about a 100,000 customers across the globe using their cloud services. They\u2019re a business cloud company that does a very good job. Raj is very knowledgeable and had been working in the cloud services environment for businesses for a long time. He\u2019s got a plan of cloud implementation that he calls the \u201cData Center Diamond,\u201d which is a holistic view of cloud migration that he\u2019s going to walk through for us today. I\u2019m really excited for you to hear Raj, and I know you\u2019ll at least get one good tidbit out of this.<\/p>\n<p>But, before we talk to Raj, I want to, again, offer you a free gift from us. It\u2019s the same gift that we talked about in our last podcast. It\u2019s a list of the best questions that you need to ask yourself about your company\u2019s needs prior to shopping public cloud service providers. This list is going to help you find the right providers fast as opposed to you finding the right providers after months of searching.<\/p>\n<p>To get this awesome list of shopping questions for Cloud VM, all you have to do is text the word \u201cCLOUDVM\u201d to 44-222. Again, text the word \u201cCLOUDVM\u201d to the number 44-222 and I will send you a free copy of all these questions that took me a long time to figure out. I, basically, had to call a ton of different cloud service providers to find out how they differ and then turn that into a list of questions, and I\u2019ll give it to you for free. Again, text the word \u201cCLOUDVM\u201d to 44-222 and I\u2019ll send you the list.<\/p>\n<p>Alright. Well, welcome to the program, Raj. Thanks for coming on.<\/p>\n<p>Raj:\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 Thank you very much for having me.<\/p>\n<p>Mike:\u00a0\u00a0\u00a0\u00a0 Great. Raj, as I mentioned, is the manager of sales engineering for Evolve IP. Raj, why don\u2019t you take a little minute to tell us a little bit about yourself both personally and professionally.<\/p>\n<p>Raj:\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 Absolutely. I\u2019ve been in the hosting industry in both the engineering and pre-sales roles for my entire career starting in 1999. Very recently joined the Evolve IP team. I live relatively nearby, grew up in the Philadelphia area, and finding it such an awesome team, like, right here in my own backyard. It, certainly, made for a great reason for me to jump over. Awesome team culture and I really enjoy being a part of this organization.<\/p>\n<p>Mike: \u00a0\u00a0 Awesome. Now, what did you do prior to Evolve IP, Raj?<\/p>\n<p>Raj:\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 I actually started way back in \u201999 as an engineer for a small hosting company up in Beltline, Pennsylvania. After a couple of years there, in 2001, I jumped over to a small company called FASTNET who was bought by a company called US LEC, who was bought by a company called PAETEC, who was a company called Windstream, and Windstream sold our division to TierPoint, the hosting division over to TierPoint<\/p>\n<p>Mike:\u00a0\u00a0\u00a0\u00a0 Okay.<\/p>\n<p>Raj:\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 Having survived five companies, four acquisitions, and a pile of t-shirts that I could swim in, I figured it\u2019s a good opportunity for me to seek out a new endeavor, and I found these guys.<\/p>\n<p>Mike: \u00a0\u00a0 Awesome. Fantastic. Tell us a little bit about yourself personally.<\/p>\n<p>Raj:\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 I\u2019m an avid hiker and kayaker. Having recently moved here, I\u2019m still trying to get myself to trying to get myself, likewise, set up a little bit. My family, my mother was from the Philippines, my father was from India. They met in Delaware, of all places. It\u2019s just, kind of, this random mix of life wise happenstance that I think has been very, very fortuitous for me.<\/p>\n<p>Mike:\u00a0\u00a0\u00a0\u00a0 Very cool. Thanks for sharing. Today, Raj is going to tell us a little bit about something he calls the Data Center Diamond. The purpose of that Data Center Diamond is to help IT professionals execute a successful cloud migration. So, Raj, I\u2019m just going to turn it over to you and let\u2019s run with it. I might stop you here and there to ask questions, but tell us about the Data Center Diamond.<\/p>\n<p>Raj:\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 Absolutely. The reason I call it the \u201cData Center Diamond\u201d is because an organization\u2019s data center is very much like a diamond. It\u2019s an expensive investment, it\u2019s unique, it\u2019s valuable, and, generally, they\u2019ve been built over a great deal of time under a great deal of pressure \u2013 much like a diamond.<\/p>\n<p>Mike: \u00a0\u00a0 Sure.<\/p>\n<p>Raj:\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 When people start thinking about cloud and looking for a provider, they intend to have a narrow focus on a specific facet if they haven\u2019t done their homework ahead of time. They\u2019re always thinking \u201cMy servers\u201d \u201cI have a list of servers.\u201d In order to successfully move in to the cloud, I feel they need to take a more holistic view and prepare a little bit before they go shopping, because when they find a provider with a list of servers, they\u2019re going to get a bunch of specs, and then, during the implementation, they\u2019re like, \u201cOh, we didn\u2019t think about this,\u201d or \u201cWe didn\u2019t recognize that,\u201d or \u201cOh, no. Our connectivity is going to take too long to bring in to that site.\u201d<\/p>\n<p>So, instead of looking at a single facet and mistaking that for the diamond, you have to look at every facet. Some things I suggest are: Looking at your network, your security, both logical and physical \u2013 What are the requirements you have for compliance? The WAN and the LAN layout \u2013 What kind of <span style=\"color: #ff6600;\"><a style=\"color: #ff6600;\" href=\"https:\/\/en.wikipedia.org\/wiki\/Subnetwork\" target=\"_blank\">subnets<\/a><\/span> do you have? Your <span style=\"color: #ff6600;\"><a style=\"color: #ff6600;\" href=\"https:\/\/en.wikipedia.org\/wiki\/Demilitarized_zone\" target=\"_blank\">DMZ<\/a><\/span> \u2013 how much bandwidth? Do you have <span style=\"color: #ff6600;\"><a style=\"color: #ff6600;\" href=\"https:\/\/www.aerocominc.com\/info\/our-products\/access\/site-to-site\/mpls\/\" target=\"_blank\">MPLS<\/a><\/span>? Are you happy with your current bandwidth provider? Things of that nature. Then, looking at disaster recovery requirements like <span style=\"color: #ff6600;\"><a style=\"color: #ff6600;\" href=\"https:\/\/en.wikipedia.org\/wiki\/Recovery_time_objective\" target=\"_blank\">RTO<\/a><\/span> and <span style=\"color: #ff6600;\"><a style=\"color: #ff6600;\" href=\"https:\/\/en.wikipedia.org\/wiki\/Recovery_point_objective\" target=\"_blank\">RPO<\/a><\/span> in the eventual disaster. And, it\u2019s another discussion completely, but when it comes to after a move, I think that most IT professionals are thinking about TTSGF. Are you aware of TTSGF?<\/p>\n<p>Mike:\u00a0\u00a0\u00a0\u00a0 No, what\u2019s that?<\/p>\n<p>Raj:\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 Well, it\u2019s, kind of, a Raj-ism but it\u2019s \u201cTime Till Somebody Gets Fired.\u201d<\/p>\n<p>Current performance issues is another thing they should be looking at. Have they been identified? Would they just be moving a problem? An example I see pretty often is that an organization has an SQL server, they\u2019re preparing for a move, and, in their physical environment or even in their virtual, they see that there is application slowness with the SQL server or things aren\u2019t running optimally. Nine times out of ten, what they do to try to resolve it is they throw resource to that (more RAM, more vCPU) and then they oversize the environment when they go to move, creating, basically, defeating the core value proposition of virtualization when the issue is actually a query that may be consuming RAMs for a selection set that were constantly updating.<\/p>\n<p>To remedy that, what I\u2019d recommend, especially for SQL servers, try taking a look at the performance monitoring log, because SQL server is typically affected by CPU, memory, file IO, or locking, blocking, or deadlocking of the processes that are being run within SQL. The performance monitor, the logs, you have to set them up just to make sure you\u2019re gathering the appropriate information.<\/p>\n<p>So, the first thing I would do is to have them run disk perf, and turn that on \u2013 that\u2019s the disk performance log within windows \u2013 then, restart the machine. Then, when you go in to the admin tool and create the performance monitoring log, I highly recommend looking at selecting the aging file processor, all SQL server counters, memory, their threads, and then looking at the system itself as well as the logical disk and physical disk.<\/p>\n<p>Being able to take that information, abstract it, will let you know whether or not it is an IO problem, or if it\u2019s a query problem, or a deadlocking issue, or CPU, so you can recognize what needs to be changed or addressed before attempting to move it, instead of just trying to move the problem, because, then, you\u2019ll just be throwing resources in the wrong way. So, I think I get a lot of good feedback from application stakeholders when they recognize that maybe they haven\u2019t been addressing the problem appropriately, but at least they have a direction they can start looking at trying to remedy before moving it to the cloud.<\/p>\n<p>Mike:\u00a0\u00a0\u00a0\u00a0 Okay. What would be some of the things that they would see, if they did those tests, that would indicate that they should move it to the cloud? I know you talked about ways they would see that they should just fix another problem and keep it in house, but what would be something they\u2019d say like, \u201cHey, this is definitely time for a new server\u201d?<\/p>\n<p>Raj:\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 Sure, absolutely. The one that I see most commonly tends to be the storage IO. Based on whatever they bought whether it be physical disks in the physical machine, or maybe they bought an undersized <span style=\"color: #ff6600;\"><a style=\"color: #ff6600;\" href=\"https:\/\/en.wikipedia.org\/wiki\/Storage_area_network\" target=\"_blank\">SAN<\/a><\/span>, or they\u2019re using <span style=\"color: #ff6600;\"><a style=\"color: #ff6600;\" href=\"https:\/\/en.wikipedia.org\/wiki\/ISCSI\" target=\"_blank\">iSCSI<\/a><\/span> instead of <span style=\"color: #ff6600;\"><a style=\"color: #ff6600;\" href=\"https:\/\/en.wikipedia.org\/wiki\/Fibre_Channel\" target=\"_blank\">fiber channel<\/a><\/span>, all these things will show an impact within that log, so they can see if there\u2019s a lot of waiting for the disk to write and commit before the next process can occur.<\/p>\n<p>Being able to \u2013 and I\u2019ll get in to a little bit later. When you\u2019re looking for a service provider, ensuring that they have the technologies that will help you alleviate that IO issue, making sure that they have appropriate speed connections to their storage, making sure that they have tiered disk levels so that you can put the SQL server on a higher tier disk so they get better IOPS out of it, and being able to recognize where the bottlenecks are. So, instead of throwing resources at it, you\u2019re sure that the infrastructure that\u2019s going to support it is going to overcome the issues based on the hardware they bought in the past.<\/p>\n<p>Mike: \u00a0\u00a0 Okay, got it. Following your model with the Data Center Diamond, what would be the next step? So, once they recognize, \u201cOkay, this is definitely something where we\u2019re going to need a new server and we want to look into going to a virtual environment or cloud environment for the server,\u201d what would be the next step in terms of preparing for that?<\/p>\n<p>Raj:\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 Alright. The first thing I would do is take it away from the technology discussion and start with the human discussion. Find the stakeholders internally and start to catalogue the applications that the organization needs to operate and survive. Figure out whether or not it can be a wholesale wreck and replace, just move the whole thing at once, or if the organization needs to move in phases based on requirements for the business.<\/p>\n<p>I had a customer who took close to a year to move their infrastructure to the cloud because they had to schedule cutovers and production stop periods based on busy periods for various departments. So, Email went first (that was easy), but Billing had to wait until after tax season, Account Payables had to wait until the end of the fiscal year, etc. Being able to recognize that there may be a timeline that suits a business to make sure that if there are issues, it doesn\u2019t impact the functional areas being moved \u2013 it certainly helps to get upfront.<\/p>\n<p>Mike:\u00a0\u00a0\u00a0\u00a0 Okay.<\/p>\n<p>Raj:\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 Once you understand the applications and their impact on the business, take a look at the servers that support them, the connectivity between them. The software, whether or not it\u2019s custom or something off the shelf \u2013 Do they work in a virtualized environment? Try to reach out and validate with the vendors. Again, things you should always do while you\u2019re considering your move to the cloud and figure out what the ramifications are.<\/p>\n<p>Then, you take that, again, the list of servers with identified dependencies and start to take a tally of their importance to the organization and their move timelines.<\/p>\n<p>Mike:\u00a0\u00a0\u00a0\u00a0 Okay. If we could rewind a second to the\u2026 We talked about the servers that support them and the connectivity, can you give us some examples of things that they would see that might affect the conversion?<\/p>\n<p>Raj:\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 Sure. I\u2019ve run into instances where we have an application server, we have a SQL server, we have a web server supporting this web-facing app for the company. The IT folks understood what these servers were and they understood, you know, \u201cAlright, if this happens, I need to restart this process,\u201d but recognizing that dependency\u2026 Like, we try to move, say, the SQL server first because they figured that was the long pole in the tent, they want to get it taken care of fast, but the latency between the SQL server and the application servers that remained within their environment in their data center, as opposed to the hosted environment, caused much more lag. So, understanding where there may be latency requirements to keep them close together so they can all be moved at once, and determine whether or not there are any other dependencies that may be calling a script from a server that wasn\u2019t identified as part of the environment during the initial phase.<\/p>\n<p>People, over time and over the course of the changes of organization and staffing, someone may have done something to band-aid an issue, and that band-aid may not translate when they try to move the operation.<\/p>\n<p>Mike:\u00a0\u00a0\u00a0\u00a0 That makes sense to even me, the non-technical\/non-IT person, especially when you\u2019re talking about latency and how the servers may be tied to one another and using one another for different things. So, that\u2019s good info.<\/p>\n<p>So, take us through the next phase. Once you\u2019ve, kind of, identified that stuff, what\u2019s next?<\/p>\n<p>Raj:\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 From there, take a look at that list and see where redundancies \u2013 I don\u2019t mean that in an infrastructure way, but I mean servers that you may not need. Where can you reduce the number of the machine count? Are there machines that are taking up resources that are retired or not in use, or just sitting off? I can\u2019t count the number of times I\u2019ve been sent a spreadsheet and every other server is actually turned off, like, \u201cOh, yeah. We just didn\u2019t clean that up,\u201d but, then, that artificially inflates the expectation for the environment on the hosted side.<\/p>\n<p>So, it\u2019s clean up time. Figure out what you really to move. Some organizations, they have redundant <span style=\"color: #ff6600;\"><a style=\"color: #ff6600;\" href=\"https:\/\/www.aerocominc.com\/info\/our-products\/cloud\/software-and-applications\/hosted-microsoft-exchange\/\" target=\"_blank\">exchange servers<\/a><\/span>. Do you really want to manage two exchange environments for that redundancy or can you replace that with something like Mindcast for email continuity?<\/p>\n<p>Once you, kind of, have an idea of that, solicit it to your stakeholder team, and you\u2019re going to argue about it. You\u2019re going to argue and argue, and refine the list, and argue about it again, and, after a few iterations, finalize what your waves are going to look like \u2013 what things are going to go, when they\u2019re going to go, and ensure that everybody agrees that this is a good path. That organization will certainly help you when you go shop for vendors as well as working on a DR strategy later down the road.<\/p>\n<p>Mike:\u00a0\u00a0\u00a0\u00a0 Okay.<\/p>\n<p>Raj:\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 Then you go looking for your vendors. You\u2019re looking for a setting for your diamond, right?<\/p>\n<p>Mike: \u00a0\u00a0 Okay. If we stop right there, how long do you think it should take from the beginning to this point in the process? So, all the internal preparation like what would be the normal \u2013 obviously, it would depend on the size of the company \u2013 but, for a normal mid-sized company with, say, like, a 1,000 users, like that type of an environment, how long do you think the process should take ideally from beginning to this point where they actually go start shopping vendors?<\/p>\n<p>Raj:\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 I think that the timeline varies based on where the direction is coming from. When it\u2019s coming from IT up, it seems like a lot of the organizations like the stakeholders don\u2019t necessarily take it seriously. They know they have to deal with it, they\u2019ll get around to it, but they\u2019re not putting something on their calendars saying \u201cWe need to sit and talk about it\u201d until someone\u2019s \u201cOh, hey, we haven\u2019t talked about this for two months,\u201d and that drags things out. When things are pushed down from the top and it becomes a mandate of participation, that shortens the timeline considerably.<\/p>\n<p>So, I\u2019ve seen people go through this process in the course of a month and I\u2019ve seen organizations go over the course in six months depending on what the organizational structure looks like, whether or not they have a project manager internally \u2013 you can, sort of, herd the cats in to a room and get them talking to each other. If it goes much beyond that, chances are the project is going to fall off the radar completely. So, I would say if you can accomplish this within two to three months, then you\u2019re in good shape.<\/p>\n<p>Mike: \u00a0\u00a0 Great. So, then take us to the next step.<\/p>\n<p>Raj:\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 Alright. Next we\u2019re going to be looking for the setting for your diamond. For a lot of those who aren\u2019t experienced with finding a cloud provider, it\u2019s a lot like buying a diamond too. They say out in the industry there are common terms like cut, clarity, color,\u00a0 carat, the four Cs of diamond buying\u2026<\/p>\n<p>Mike:\u00a0\u00a0\u00a0\u00a0 Oh, man. You are so taking me back when I was engaged or prior to getting engaged, and looking at that looking glass all the time and all the little diamond vendors. Man, it was like I knew that thing inside and out. I think I researched on the internet how to buy a diamond. Man, yeah, I know exactly what you\u2019re talking about.<\/p>\n<p>Raj:\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 Yeah. So, when you\u2019re walking in, you have no idea what these terms mean and every vendor is going to give you a similar definition, but they all have their own nuances.<\/p>\n<p>I\u2019ll kind of throw a couple of potential pitfalls out there, things to look out for. So, compliance, right? Today, in the hosting industry, <span style=\"color: #ff6600;\"><a style=\"color: #ff6600;\" href=\"https:\/\/en.wikipedia.org\/wiki\/SSAE_16\" target=\"_blank\">SSAE 16<\/a><\/span>, which is like an audit, SOC-2 (Statement of Compliance type 2), which means the compliance checks are being done over the course of time is considered table stakes for a hosting organization \u2013 so, as a bare minimum \u2013 but not all SSAE 16 reports are the same. So, see what the report itself\u2026 The organization creates a list of controls: Do we have biometrics? Do we have sign-ins? Do we have cameras on site? Do we do generator maintenance? Things like that \u2013 controls that can assure that the company is, like, maintaining and managing up time and availability.<\/p>\n<p>But, not all these controls are the same. One company may say, \u201cYes, we have three cameras and they are recorded to DVR for 24 hours,\u201d another organization may say, \u201cWe have cameras in every entrance, exit, and throughout the data center floor and we\u2019ve maintained those for three years,\u201d or whatever the case may be. So, see if you can ask where \u2013 if they\u2019ll release it, possibly, on your NDA, find out if the vendor that you\u2019re looking at will provide you with a list of the controls. Then, ask for a copy of the auditor\u2019s opinion letter, so you can take a look to see what they consider important to provide you the service that they provide, and the third-party auditor can provide an opinion \u201cYes, they are meeting the criteria that they set forth in these controls or they\u2019re not \u2013 that can help provide some insight as to whether or not they\u2019re the right place to put your data.<\/p>\n<p>Then, I\u2019m looking at what other service that they can provide to me to the requirements of any other compliance that you may have, so <span style=\"color: #ff6600;\"><a style=\"color: #ff6600;\" href=\"https:\/\/en.wikipedia.org\/wiki\/Health_Insurance_Portability_and_Accountability_Act\" target=\"_blank\">HIPAA<\/a><\/span>, <span style=\"color: #ff6600;\"><a style=\"color: #ff6600;\" href=\"https:\/\/en.wikipedia.org\/wiki\/Payment_Card_Industry_Data_Security_Standard\" target=\"_blank\">PCI<\/a><\/span> compliance. Can they provide you things like dedicated firewalls or dedicated servers for data that cannot share residence with other customers? Can they provide you encrypted storage? These are all great things to ask upfront and help to vet them in or out of the list as far as the solution.<\/p>\n<p>Then, looking at the infrastructure. What are their CoD\/PoDs built on? Is it proprietary Hypervisor or design, or is it something industry standard like VMWare or Hyper-V. At least you know whether or not you\u2019re going to have difficulty moving your environment out again in the event that you decide to migrate sometime in the future. For instance, like some providers using newer Hyper-converge, they found it sexy, they bought it, now they\u2019re like, \u201cI can\u2019t move these stuff,\u201d or something like <span style=\"color: #ff6600;\"><a style=\"color: #ff6600;\" href=\"http:\/\/www.aerocominc.com\/company-profile\/amazon-web-services\" target=\"_blank\">AWS<\/a><\/span>. I always, kind of, compare those situations like the Hotel California because it\u2019s a lovely place, but you can\u2019t ever leave.<\/p>\n<p>Mike:\u00a0\u00a0\u00a0\u00a0 Very true.<\/p>\n<p>Raj: \u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 If you identify whether their platform will actually help those performance issues, you identified in the earlier analysis like the IO to the storage platform, backplane connectivity of the underlying infrastructure, whether it be 1 gig, or 10 gig, or redundant 10gig, determine their storage connectivity. If they can provide you that tiered storage so your faster needing apps can be on the higher speed storage\u2026 You can ask for things like if they can provide you with an idea of the anticipated IOPS to compare against what you\u2019ve identified from your performance metrics up top.<\/p>\n<p>Then, of course, asking about the business stuff, SLAs and managed services. Can they provide you the SLAs? Can you augment your staff and free up valuable staff time so they can pay attention to your core business applications or end users instead of doing patch management at 4:00 a.m? What about monitoring alert? When you\u2019re on vacation, do you want to be worried about whether or not your service is up or down as opposed to if you want the back of your beach chair to be up or down, right? Let someone else provide you that coverage.<\/p>\n<p>Mike:\u00a0\u00a0\u00a0\u00a0 In terms of that, what would be an example of what they might see that is an alert to them that, hey, they\u2019re going to have to be involved all the time? What type of notifications should they be looking for if they don\u2019t want to be involved all the time? Because, I know I\u2019ve been in a lot of these meetings, and everybody says they give some level of notification, and you kind of have to usually dig down with that provider to figure out that, \u201cYou know what? This isn\u2019t much notification at all. I\u2019m going to have to be involved in this.\u201d<\/p>\n<p>Raj: \u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 Largely, usually, monitoring tends to become a managed service, which is an add-on to basic infrastructure as a service. Some cloud providers do offer it, include it, as part of their packages, but most of them are just doing up\/down. Is the server up or is the server down? We\u2019ll send you an email if it\u2019s down. Some of them will call you when it\u2019s down, but then, you have deeper levels of monitoring available to you. So, is my CPU utilization too high? Is my memory utilization too high? Are my drives starting to get full and past a certain threshold? They\u2019ll alert you on those things, not just whether or not it\u2019s alive, because in health VS life, your life is going to be shortened if you\u2019re not keeping healthy.<\/p>\n<p>Mike:\u00a0\u00a0\u00a0\u00a0 Yeah. What about SLAs? You mentioned that too. What should an IT buyer be looking for in terms of SLAs on their hosted environment? Because I know everybody is got them and some people balk at SLAs and say they\u2019re not worth much, but what are some things that are important on SLAs that they should be really looking for?<\/p>\n<p>Raj:\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 Well, most of all that the SLAs have teeth, that there is a mediation process and that the remuneration for it is actually something useful to them. Some will say, \u201cOh, yeah, we\u2019ll give you back however long the system was unavailable beyond our SLA,\u201d and that could be minutes, so that doesn\u2019t make sense. Other organizations may say, \u201cWe\u2019ll credit you a free full day service if we hadn\u2019t met our SLA.\u201d So, looking through these definitions as far as what their metrics are for measuring SLA success, sort of, the recompensation to the customer, and whether or not there are any exclusionary or indemnifications that may allow the hosting provider to just say, \u201cWell, you know, we didn\u2019t meet our SLA, but we haven\u2019t done it three months in a row, so we\u2019re not doing anything for you.\u201d<\/p>\n<p>Mike:\u00a0\u00a0\u00a0\u00a0 Right.<\/p>\n<p>Raj:\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 Just be very careful of that language, running it through your\u2026 If you have a legal team or if you have a contracts team, make sure they understand exactly what the SLA says to them because some people\u2026 Look, I mean, for a long time, the hyperscalers, they offer a great SLA, but zero compensation and, in the event of a failure, zero recompence for the company. So, the north east grid goes down three times in a month, it\u2019s like, \u201cSorry. It\u2019s in our SLA. You should have read it.\u201d<\/p>\n<p>Mike:\u00a0\u00a0\u00a0\u00a0 Or, it\u2019s even worse. They\u2019ll say oh, for, you know, so you were down for four hours during the middle of the day that day, \u201cWell, you know, we\u2019ll credit you back for those four hours.\u201d So, four hours divided by the number of hours in a month, you end up getting a $100 credit.<\/p>\n<p>Raj:\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 Yeah.<\/p>\n<p>Mike:\u00a0\u00a0\u00a0\u00a0 Like we\u2019re down for four hours, they are probably business hours, and we get a $100 back to our company doesn\u2019t mean much.<\/p>\n<p>Raj:\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 Exactly. And, making sure that you understand what that means to you. Some of that may require looking at, \u201cAlright, how do we further mitigate the risk of possible downtime?\u201d<\/p>\n<p>Mike:\u00a0\u00a0\u00a0\u00a0 Right.<\/p>\n<p>Raj:\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 The last thing that I have is about the move itself. If you have a telecom vendor of choice, make sure you can find out ahead of time, once you\u2019ve identified your hosting vendors, find out if the vendor is on-net If they are, start looking at the timelines for getting necessary connections deployed to the site [where your infrastructure will be homed because this can greatly affect your move timelines. If you don\u2019t want to be blindsided by additional costs or delays after you\u2019ve signed for that contract&#8230; Because, I\u2019ve seen instances where someone will sign the contract and they\u2019re thinking, \u201cAlright, we can get started immediately,\u201d but they record an <span style=\"color: #ff6600;\"><a style=\"color: #ff6600;\" href=\"https:\/\/www.aerocominc.com\/info\/our-products\/access\/site-to-site\/mpls\/\" target=\"_blank\">MPLS<\/a><\/span> drop and they only looked at that after they selected the vendor.<\/p>\n<p>The \u00a0MPLS vendor could build in, sure, but based on their current timelines and then a major telco strike, that went from a sixty-day install to a 180-day install which means that the customer is now paying for infrastructure that they can\u2019t access. However, some hosting providers will provide alternate methods \u2013 they can always get started. Possibly drop a site-to-site VPN in place over open internet which will at least allow them to get access to the environment and start their bill while waiting for the MPLS to build in.<\/p>\n<p>Mike:\u00a0\u00a0\u00a0\u00a0 Cool. So, that\u2019s the Data Center Diamond?<\/p>\n<p>Raj:\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 That\u2019s the Data Center Diamond.<\/p>\n<p>Mike: \u00a0\u00a0 Thanks for sharing that. I think that was some good quick tips that I think all the audience can, at least, pull something out of to help make their next install go a little bit better. I appreciate you sharing it with us. One thing I definitely want to do is learn a little bit about Evolve IP, but before we go there, I always like to ask our guests to tell us a little funny story or entertaining story about what\u2019s the most interesting, or funny, or peculiar thing you\u2019ve ever experienced at work?<\/p>\n<p>Raj:\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 Oh, absolutely. So, over the course of the years, being in, sort of, mixed environments of telecom and hosting, there\u2019s often opportunities where someone like the telco rep and the telco SE will go out with the hosting rep and the hosting SE to do a joint meeting with the potential client.<\/p>\n<p>I was in a meeting in New York City and the four of us we were sitting in front of the client, and\u2026 This actually has nothing to do with technology, but about language. The telco rep went first and he kept on saying, \u201cWe can build this on-premise,\u201d \u201cWe can do this on premise,\u201d \u201cWe can monitor this on premise.\u201d After about ten minutes of him speaking, the client slams his fist down on the table and says \u201cStop. My father was a English teacher for 25 years and he thought me that words have meanings. A \u2018premise\u2019 is an insertion or a proposition which forms the basis for an undertaking statement or theory. A \u2018premises\u2019 is a tract of land with a building on it. You sound like an idiot, get out of my office,\u201d and he sent the telco reps out. He, liked, turned to us and started talking hosting.<\/p>\n<p>Ever since that moment I realized that \u201con-premise\u201d \u2013 we\u2019ve lost the language form. \u201cOn-premise\u201d has become a telco mainstream phrase that every time I see it, read it here, I think of that moment, and I\u2019m very conscious about using \u201cpremises\u201d when speaking about a location.<\/p>\n<p>Mike:\u00a0\u00a0\u00a0\u00a0 Oh, my God. That\u2019s crazy. That\u2019s hilarious. You know what? I\u2019ve been on the online forums, I think, on Spiceworks and Reddit. Seeing that in the IT forums where somebody will say \u201con-premise\u201d and then somebody always jumps in and corrects them, like, gives those two definitions. I always thought it was, kind of, like everybody understood that it\u2019s just a slip of the tongue and, like you said, it\u2019s just, kind of, evolved over time to mean something that it really isn\u2019t in terms of the word. But, they take it seriously. They kicked the rep out of the office.<\/p>\n<p>Raj:\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 Kicked the rep out of the office.<\/p>\n<p>Mike:\u00a0\u00a0\u00a0\u00a0 Yeah. That had to be an uncomfortable minute or so \u2013 everybody standing there like, \u201cOkay. Alright. We\u2019re going to keep going.\u201d<\/p>\n<p>Raj:\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 Yeah.<\/p>\n<p>Mike:\u00a0\u00a0\u00a0\u00a0 Oh, man. So, like, \u201cYeah, can we swap reps here or something? What\u2019s going on?\u201d That\u2019s hilarious. That\u2019s awesome. Thanks for sharing that.<\/p>\n<p>Alright. Well, tell us a little bit about who is <span style=\"color: #ff6600;\"><a style=\"color: #ff6600;\" href=\"http:\/\/www.aerocominc.com\/company-profile\/evolve-ip\" target=\"_blank\">Evolve IP<\/a><\/span> and what you guys have going on that\u2019s exciting. Obviously, we\u2019ve heard from you, you\u2019ve told us some really good stuff, you obviously know what you\u2019re talking about, but there\u2019s a company you work for and, obviously, it\u2019s a company that you felt is doing a good job, so we\u2019d like to hear about it.<\/p>\n<p>Raj:\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 Oh, absolutely. Evolve IP, headquartered in Wayne, Pennsylvania with data centers in Las Vegas and Austin, Texas, and teams spread throughout the country. We\u2019re an award-winning cloud services company with a fantastic portfolio including virtual data centers, <span style=\"color: #ff6600;\"><a style=\"color: #ff6600;\" href=\"https:\/\/www.aerocominc.com\/info\/our-products\/cloud\/servers-desktop-infrastructure\/public-shared-cloud-servers\/\" target=\"_blank\">infrastructure as a service<\/a><\/span>, <span style=\"color: #ff6600;\"><a style=\"color: #ff6600;\" href=\"https:\/\/www.aerocominc.com\/info\/our-products\/cloud\/backup-and-security\/data-disaster-recovery\/\" target=\"_blank\">disaster recovery as a service<\/a><\/span>, <span style=\"color: #ff6600;\"><a style=\"color: #ff6600;\" href=\"https:\/\/www.aerocominc.com\/info\/our-products\/cloud\/servers-desktop-infrastructure\/desktop-as-a-service\/\" target=\"_blank\">virtual desktops<\/a><\/span>, <span style=\"color: #ff6600;\"><a style=\"color: #ff6600;\" href=\"https:\/\/www.aerocominc.com\/info\/our-products\/cloud\/communications-and-collaboration\/cloud-phone-system\/\" target=\"_blank\">IP phone systems<\/a><\/span>, and <span style=\"color: #ff6600;\"><a style=\"color: #ff6600;\" href=\"https:\/\/www.aerocominc.com\/info\/our-products\/cloud\/communications-and-collaboration\/cloud-call-center\/\" target=\"_blank\">call center solutions<\/a><\/span>. We also offer IT as a service like <span style=\"color: #ff6600;\"><a style=\"color: #ff6600;\" href=\"https:\/\/www.aerocominc.com\/info\/product\/it-support\/outsourced-help-desk-support\/\" target=\"_blank\">workstation and server helpdesk<\/a><\/span> for end users, and domestic-based O365 support. In the U.S, we\u2019ve got something exciting going on with Evolve IP.<\/p>\n<p>There are a couple of things that I love about this company. One, we are very committed to giving back to the community. We support over 45 charities in the last couple of years, raised over $4 million for those charities. Also, as mark of our recent success, a private equity firm has recently bought in to evolve IP and have created a $100 million war chest for us to continue expansion of services, products, staff, and strategic acquisitions.<\/p>\n<p>Mike:\u00a0\u00a0\u00a0\u00a0 Wow, that\u2019s great.<\/p>\n<p>Raj:\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 There\u2019s a lot going on here. So, I\u2019m very excited to be part of Evolve IP.<\/p>\n<p>Mike:\u00a0\u00a0\u00a0\u00a0 Yeah. That\u2019s a nice chunk of money for you guys to do something with. That\u2019s cool. So, obviously, you mentioned that you\u2019re fairly new to Evolve IP. When you were looking at Evolve IP, you\u2019ve been around the block, you\u2019ve seen a couple of different companies. In the time that you\u2019ve been in Evolve, what have you seen that you really like that you think makes you guys a little different than the other cloud and telecom service providers out there?<\/p>\n<p>Raj:\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 Oh, I think the main thing is that we\u2019re nimble. A lot of the smaller cloud providers, like where I started out once upon a time, has a very similar entrepreneurial feeling to that. However, being eaten by telcos over the years, I found that a lot of times the hosting kind of dilutes and becomes , you know, starts falling into telco processes. That makes it a little bit more difficult to react quickly to changes in technology, changes in the market place because that has to be codified through the parent company.<\/p>\n<p>As a nimble, agile organization with a deep bench\u2026 I mean, our leadership team has over 27 years combined experience both in the telco side and hosting, and an organization where we are looking for the right kind of talent, and then finding where in the organization they can fit in to make the clock run better.<\/p>\n<p>Mike:\u00a0\u00a0\u00a0\u00a0 Yeah. Personally, I\u2019ve worked with Evolve IP for at least a couple of years now and I found the same thing where it\u2019s got some of the telco elements in it just in terms of the product set, but it\u2019s a very cloud-driven company and you guys have got a lot of cool things that you do in terms of the cloud product set. I mean, we\u2019ve quoted you guys for, I think, hosted VoIP for a couple of providers, kind of, diving in a little bit in to the telco side, but the cloud side is also really cool in terms of what you guys are able to do and how flexible you guys are with working with certain customer requirements. I agree with you \u2013 the flexibility is great.<\/p>\n<p>Well, hey, that\u2019s about it. I appreciate you jumping on the show with us and taking the time. I know that you\u2019re super busy. Sales people are always pulling in different directions to meet with with customers for you to take an hour out of your time to sit and chat with us. I really appreciate it and I hope that it\u2019s been as valuable to our listeners as it has been for me. Thanks a lot, Raj. I appreciate it.<\/p>\n<p>Raj:\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 You\u2019re very welcome. Thank you, again, for having me on. I look forward to hearing future podcasts from you guys.<\/p>\n<p>Mike: \u00a0\u00a0 Absolutely. Take care.<\/p>\n<p>Raj:\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 Take care and have a great day.<\/p>\n<p>Mike: \u00a0\u00a0 So, that was Raj from Evolve IP. He\u2019s awesome, isn\u2019t he? Before I let you go, I just wanted to give you one more quick reminder about our free gift. It\u2019s all the best questions you should ask yourself before shopping for Cloud VMs. All you have to do to get it is text the word \u201cCLOUDVM\u201d to the number 44-222. Again, text the word \u201cCLOUDVM\u201d to the number 44-222, and we\u2019ll send you that list of questions. Have a great day.<\/p>\n<p>&nbsp;<\/p>\n","protected":false},"excerpt":{"rendered":"<p>In this episode, Evolve IP&#8216;s Manager of Sales Engineering, Raj Palat, outlines his insightful &#8220;Data Center Diamond&#8221; approach to cloud implementation. It&#8217;s a holistic approach to evaluating your cloud decision from every angle, which is exactly what your boss will<span class=\"ellipsis\">&hellip;<\/span> <a href=\"https:\/\/www.aerocominc.com\/info\/cloud-therapy-ep-005-improving-your-cloud-server-implementation-with-raj-palat\/\"><\/p>\n<div class=\"read-more\">Read more &#8250;<\/div>\n<p><!-- end of .read-more --><\/a><\/p>\n","protected":false},"author":6,"featured_media":9752,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"om_disable_all_campaigns":false,"_mi_skip_tracking":false,"_monsterinsights_sitenote_active":false,"_monsterinsights_sitenote_note":"","_monsterinsights_sitenote_category":0},"categories":[371],"tags":[560,575,574,422,667,429,436,684,537,680,689,688,691,438,681,561,687,176,367,685,479,692,423,437,651,480,679,683,682,686,693,690],"_links":{"self":[{"href":"https:\/\/www.aerocominc.com\/info\/wp-json\/wp\/v2\/posts\/9496"}],"collection":[{"href":"https:\/\/www.aerocominc.com\/info\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.aerocominc.com\/info\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.aerocominc.com\/info\/wp-json\/wp\/v2\/users\/6"}],"replies":[{"embeddable":true,"href":"https:\/\/www.aerocominc.com\/info\/wp-json\/wp\/v2\/comments?post=9496"}],"version-history":[{"count":0,"href":"https:\/\/www.aerocominc.com\/info\/wp-json\/wp\/v2\/posts\/9496\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.aerocominc.com\/info\/wp-json\/wp\/v2\/media\/9752"}],"wp:attachment":[{"href":"https:\/\/www.aerocominc.com\/info\/wp-json\/wp\/v2\/media?parent=9496"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.aerocominc.com\/info\/wp-json\/wp\/v2\/categories?post=9496"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.aerocominc.com\/info\/wp-json\/wp\/v2\/tags?post=9496"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}