Recent Articles
Hadoop- Metadata Scalability
As the cluster grows, the metadata requires more memory. One petabyte of storage requires one gigabyte of memory. Discovering data across different systems becomes a significant challenge.
Big Data I/O Forum Approach
Big Data applications are proliferating rapidly, our objective will be critical to customer’s successful implementation of Big Data applications. The Forum will explore in-dept issues impacting their application, identify real-world paradigms and provide solutions. The Forum will reach out and educate the community and help accelerate adoption of new technologies.
UCSD: Gordon tackling Data Intensive Applications
Data intensive applications are creating a data tsunami requiring new architecture and new ideas. The good folks down at USCD have received a $20 million grant form NSF to attack the data intensive applications. Gordan, AKA Flash Gordon is a supercomputer based on SSD Flash Memory and Virtual Shared Memory.
Digital Intermediate
Avatar has moved the bar for digitizing motion pictures and manipulating color and adding dimensions and image characteristics. This has created a new bottleneck in storage access and final creation adjustment to deliver final movies to theaters.
Lustre Bottlenecks
When you upgrade new processors into a Lustre environment of petabytes of storage, the Beowulf roars with indigestion.
Boot Storms
Boot Storms a new challenge in enterprise VMware environments. Every Monday morning we boot up our VDI desktops and the Tsunami hits the storage farm!
