![\"donald](\"https:\/\/i0.wp.com\/timoelliott.com\/blog\/wp-content\/uploads\/2013\/04\/donald-feinberg-banner.jpg?resize=690%2C310&ssl=1\" "\\"donald")
<\/p>\n<\/p>\n
I attended a SAP-sponsored breakfast meeting organized in Sydney earlier this year, as part of a\u00a0ANZ\/APJ innovation analytics tour<\/a>. The meeting was hosted by Ryan Blackwood<\/a>, Director of Strategy & Innovation for SAP Australia, and featured a talk by Donald Feinberg, Gartner VP and Distinguished Analyst<\/a> on the \u201cNexus of Forces<\/a>\u201d: social, mobile, cloud and information.<\/p>\n Here the first part of the notes I took from Donald’s presentation, formatted for legibility. In Part 2, Donald talks about the business impact of in-memory technologies<\/a>\u00a0and in Part 3,\u00a0how to create an in-memory action plan<\/a>.<\/p>\n Some people think that we\u2019re talking about in-memory because of all the hype that\u2019s been going on in the market \u2013 but it\u2019s not just SAP. Every single vendor out there is getting involved in in-memory technology. We actually started looking at in-memory technology at Gartner back in 2009 and in-memory databases have been around for years in some form going back to the 90s.<\/p>\n This technology is now becoming disruptive.<\/strong><\/p>\n Myths:<\/p>\n In-memory computing will have a long term, disruptive impact by radically changing users\u2019 expectations, application design principles, products\u2019 architecture and vendors\u2019 strategy<\/p><\/blockquote>\n This is going to change the way you do everything. Everybody in this room will be running their entire IT organization in-memory in the next 15 to 20 years<\/strong>. It\u2019s not going to happen overnight. But within the next 15 years, you will run your whole operation in memory. You won\u2019t have tape drives, you won\u2019t have disk drives. You\u2019ll be using flash and memory. Flash will be your backup, your archive, and memory is where you\u2019re going to run everything. And that\u2019s absolutely a fact.<\/p>\n When we talk about in-memory computing, we are talking about DRAM: the \u201cd\u201d stands for destructive: it doesn\u2019t hold data it if you lose power. It\u2019s not about flash or NAND memory. Flash is a form of memory, but it\u2019s not what we\u2019re talking about when we talk about in-memory computing.<\/p>\n All forms of flash today are used like disk drives. Even though we may remove the controller as a bottleneck, the applications are still doing I\/O to a flash drive or a flash board. It is getting much more reliable and cheaper, so it is going to become a persistence mechanism replacing disk.<\/p>\n Today, the reliability of flash is longer than that of disk drives. If you replace your hardware every three to four years, and you have flash SSD and disk, you will probably not see a failure on the flash at all in that period of time, but I guarantee you that you will change disk drives.<\/p>\n When we talk about in-memory, we are talking about the physical database being in-memory rather than as it is \u201ctraditionally\u201d done: on disk.<\/p>\n What is the difference? Database engines today do I\/O. So if they want to get a record, they read. If they want to write a record, they write, update, delete, etc. The application, which in this case is a DBMS, thinks that it\u2019s always writing to disk. If that record that they\u2019re reading and writing happens to be in flash, it will certainly be faster, but it\u2019s still reading and writing. Even if I\u2019ve cached it in DRAM, it\u2019s the same thing: I\u2019m still reading and writing.<\/p>\n What we\u2019re talking about here is the actual database is physically in in-memory. I\u2019m doing a fetch to get data and not a read. So the logic of the database changes. That\u2019s what in-memory is about as opposed to the traditional types of computing.<\/p>\n Why now? The most important thing is this: DRAM costs are dropping about 32% every 12 months. Things are getting bigger, and costs are getting lower. If you looked at the price of a Dell server with a terabyte of memory three years ago, it was almost $100,000 on their internet site. Today, a server with more cores — sixteen instead of twelve — and a terabyte of DRAM, costs less than $40,000.<\/p>\n So the costs of this stuff is not outrageous. For those of you who don\u2019t understand storage, I always get into this argument: the total cost of acquisition of an in-memory system is likely higher than a storage system. There\u2019s no question. But the total cost of TCO is lower \u2013 because you don\u2019t need storage people to manage memory. There are no LUNs [logical unit numbers]: all the things your storage technicians do goes away.<\/p>\n People cost more than hardware and software \u2013 a lot more. So the TCO is lower. And also, by the way, power: one study IBM did showed that memory is 99% less power than spinning disks. So unless you happen to be an electric company, that\u2019s going to mean a lot to you. Cooling is lower, everything is lower.<\/p>\n So don\u2019t let somebody say to you we can\u2019t go in-memory because it\u2019s so much more money. Acquisition costs may be higher. If you calculate out a TCO, it\u2019s going to be less.<\/strong><\/p>\n >>Continue to Part 2, where\u00a0Donald\u00a0talks about the\u00a0business impact of in-memory technologies<\/a>, or\u00a0Part 3,\u00a0how to create an in-memory action plan<\/a>.<\/p>\n
\nCommon in-memory myths<\/h3>\n
![\"in-memory-myths[3]\"](\"https:\/\/i0.wp.com\/timoelliott.com\/blog\/wp-content\/uploads\/2013\/04\/in-memory-myths3_thumb.jpg?resize=571%2C441&ssl=1\" "\\"in-memory-myths[3]\\"")
<\/a><\/p>\n\n
In-memory is going to change the way you do everything<\/h3>\n
What is in-memory computing?<\/h3>\n
![\"what-is-in-memory\"](\"https:\/\/i0.wp.com\/timoelliott.com\/blog\/wp-content\/uploads\/2013\/04\/what-is-in-memory_thumb.jpg?resize=538%2C396&ssl=1\" "\\"what-is-in-memory\\"")
<\/a><\/p>\nWhy is it time for in-memory computing?<\/h3>\n
In-memory results in lower total cost of ownership<\/h3>\n