Sean McKeown, Cisco | Big Data New York City 2016
summary
Cisco moves CPU closer to data and designs horizontally scalable clusters that take a five‑node POC to 500+ servers and even thousands of nodes in production. Hardware + ISV validated designs with Cloudera Hortonworks MapR IBM and Splunk use UCS service profiles to templatize firmware and network settings and UCS Director Express for Big Data to provision ISV tools. A new c3260 4RU chassis lets sites add drives or a second server to change compute/storage ratios, while TPCx‑HS benchmarks and 10G→40G network needs expose real bottlenecks.
AI-written summary · about
chapters (11)
- 0:00Start
- 0:2101. Sean McKeown, Cisco, Visits theCUBE!.
- 1:0102. Talk About Your Role And How It Relates To Big Data.
- 2:4903. How Do The Requirements Of Big Data Differ From An Architectural Perspective.
- 4:5004. What Is Cisco's Role Here.
- 6:5905. Do You Remove A Lot Of The Complexity.
- 8:3906. Do People Complain About Being Able To Scale To Compute Independant From Storage.
- 10:2307. Discuss ISV's Role.
- 12:1008. What's The Gnarliest Big Data Problem From An Architectural Aspect.
- 14:3809. Do We Need Fast Pipes Between Servers Now.
- 16:1210. What Do You Want People To Know About Cisco And Big Data.
