Paper Title: Persistent Memory: The Answer to Today's Data Center Challenges

Paper Abstract: Persistent memory, such as 3D XPoint or MRAM, offers close to DRAM speeds (and latency) at close to flash prices. Its use will lead to far better performance for cognitive computing, in-memory databases (such as Spark, Hadoop, and HANA), deep learning, pattern and real-time fraud detection, artificial intelligence, real-time analytics, data reduction, and cybersecurity. System and software developers will no longer have to work around relatively slow drive accesses, or worry about long latencies caused by data transfers that end up involving flash memory rather than DRAM. Of course, the transition will not be without its own challenges. Storage architects must deal with slower than DRAM speeds and endurance issues, and must develop innovative approaches including tiers and caches. Making effective use of persistent memory will take time, and will require advances in hardware, software, and applications.

Paper Author: Andy Walls, Fellow/CTO/Chief Architect, IBM

Author Bio: Andy Walls is Chief Architect and CTO for IBM’s Flash Systems Division. He is also an IBM Fellow, the company’s most prestigious honor. A 35-year storage industry veteran, Andy is a pioneer in enabling flash memory in the enterprise. He has developed enterprise storage systems that achieve high performance, good endurance, and the availability data centers require. He was responsible for the Texas Memory Systems acquisition and has since defined the architecture for all FlashSystem products. He is currently defining next generation products that can be used in traditional SAN environments as well as in clouds and by emerging workloads. He is widely recognized an expert in storage and flash memory. He has designed ASICs, PCBs, firmware stacks, and systems. Known as an innovator, he has filed over 100 patents. Andy earned a BSEE from UC Santa Barbara.