Presentation Transcript: Best Practices around How to Economically Boost System and Application Performance with NAND Flash
NAND flash is changing enterprise storage architectures now. Read this transcript and learn how to optimize your storage infrastructure using the latest NAND technologies, HDD and DRAM. We'll discuss application and performance metrics, addressing of disk drives and solid state drives (SSDs), endurance, reliability and the cost saving opportunities that flash presents in today's datacenter. We'll also explain how you can utilize flash technology and the Solaris ZFS™ file system to take advantage of high performance enterprise SSDs and the low cost of high-capacity hard disks to create hybrid storage pool solutions that help balance system performance and cost.
Optimizing CPU/Controller and Storage Systems
Slower storage devices can constrain today's multicore, multithreaded application server systems. When requesting data, servers frequently wait for storage while application performance remains sluggish, regardless of CPU horsepower. The traditional remedy of adding expensive DRAM may no longer suffice as data sets double every two years.
It's also well known that Moore's Law is outstripping disk drive performance. As a result, servers and storage systems are unbalanced between CPU/controller capability and storage pool performance. The objective of modern system design is to rebalance the CPU-storage ecosystem while optimizing both low $/GB and $/IOPS. In this mediacast, we'll compare the cost, power consumption, random I/O and latency for DRAM, server flash and HDD.