FlexMem Bridge Technology-The Keyword is Flex-ibility

I almost called this post "Flexing your memory muscle, not your data center footprint." Recently, some of our competitors have been working hard trying to discredit Dell's FlexMem Bridge Technology. We get it; it's what most vendors do when they don't understand a new technology. Dell's approach to maximizing ROI on memory technology investments is unique.

Instead of forcing customers into higher priced software licensing bands due to memory constraints or giving them lots of memory in larger form factors that eat away their precious data center space, we innovated and delivered FlexMem Bridge technology. The beauty of this technology is that it solves customers' memory-bottleneck issues and rack-density concerns by allowing them to stay in the same software license cost scheme and get 32 DIMMs with two processors in a slim 2U form factor. A concept truly inspired by those living in the real world. When the customer is ready for 4 processors, they simply replace the FlexMem Bridge with new processors and instantly double processor performance-all without taking up additional space!

To further set the record straight, I'd like to highlight some key points (competitors should take notes):

  • FlexMem Bridge does not affect memory speed. With FlexMem Bridge, the PowerEdge R810 and M910 full-height blade can support up to 1066MHz.
  • This feature was designed by customer feedback and need.e listened to customers and delivered.
  • FlemMem Bridge utilizes two memory controllers per processor, not one Check out this whitepaper PDF file for more detail.

Now, I want to focus on what our competitors aren't telling customers:

  • The R810 delivers up to 4 processors and 32 DIMMs in an efficient 2U form factor-previously the only option for customers was a 4-socket capable server in a 4U form factor. Customer were forced to cut rack density in half; that's not the case with R810.
  • The R810 delivers double the memory capacity at 512GB and double the processing power of any previous generation 4-socket, 4U server.
  • The flexibility of the R810 can serve the needs of memory-constrained 2 socket, 2U servers in the same form factor and deliver 166% more memory than Intel based 2 socket platforms. This is also an ideal platform for the current 4 socket, 4U customer that still needs performance, but needs to shrink their data center footprint.
  • We don't force customers to buy any additional memory modules that eat up rack density; we can scale memory internally.
  • Customers don't buy servers only for memory, they buy servers that can maximize their workloads, maximize performance per watt and have a well-balanced thermal footprint. Applications need a combination of processor, memory, and I/O.Vendors can't over index on memory for the sake of rack density and processing density

I'm proud of Dell's FlexMem Bridge technology because it illustrates how we listen to our customers, helping them solve real world challenges.

Don't take my word for it… here's what the industry is also saying:

"Dell has introduced patent-pending FlexMemory technology specifically for the Nehalem-EX servers, which can scale the memory capacity without adding new hardware. The new technology is silicon that manages and manipulates the memory in order to double capacity, overcoming limitations that typically throttle the amount of memory that can fit in servers."  Agam Shah, IDG News Service

"We don't see any real competition for the newest Intel Xeon when running ERP, OLTP or virtualization loads. Dell's R810 has made these kinds of high reliability servers more accessible with the R810, and for that we have to congratulate them."  Johan De Galas, Anandtech

Hardware.info gave the PowerEdge R810 an excellent review, stating "Dell cunningly adapts to Intel's strategy, which recommends that the processors in this segment are for use in dual-CPU configurations, and thanks to the special socket adapters, all memory slots can be used. This makes the platform also fit for workloads where there is more demand for memory instead of computing power."

About the Author: Armando Acosta

Armando Acosta has been involved in the IT Industry over the last 15 years with experience in architecting IT solutions and product-marketing, management, planning, and strategy. Armando’s latest role has been focused on Big Data|Hadoop solutions, addressing solutions that build new capabilities for emerging customer needs, and assists with the roadmap for new products and features. Armando is a graduate of University of Texas at Austin and resides in Austin, TX.