Showing posts with label EMC. Show all posts
Showing posts with label EMC. Show all posts

Monday, February 6, 2012

VMware Global Alliances Blog: EMC & VMware: Get Maximum Performance and Availability for Virtualized Applications

VMware Global Alliances Blog: EMC & VMware: Get Maximum Performance and Availability for Virtualized Applications:

Kannanm
Posted by Kannan Mani
Technical Solutions Architect


Check out the video below where Jason Kosaftis from EMC and I discuss on the upcoming joint webcast series - virtualizing Oracle databases. The combination of VMware vCenter™ and EMC FAST Cache dynamically provisions server and storage resources to allow maximum performance at scale in virtualized Oracle environments. With the latest release of VMware vSphere® and the fully automated Storage Tiering capabilities of EMC VNX Unified Storage, Oracle DBAs and IT infrastructure teams can now accelerate server and storage resource provisioning and minimize database performance tuning while maintaining maximum availability.

During these webcasts you will hear from EMC and VMware Engineers, as well as end customers and IT executives from some of our largest customers discuss on cost savings, performance and efficiency they have achieved using VMware and EMC technologies. Please visit the links below and register today.



Maximum Performance and Availability for Virtualized Oracle Databases with VMware and EMC
Hosts: George Trujillo, Tier One Database Specialist, VMware and Sam Lucido, Solutions Consultant, EMC

Thursday, February 16, 2012, 8:00 AM PST
Register today!


Tuesday, February 28, 2012, 9:00 AM PST
Register today!

Next up from EMC: Project Thunder flash appliance

Next up from EMC: Project Thunder flash appliance:


EMC’s promised Project Lightning server-based flash storage is now available — under the VFCache brand. But EMC’s not done — it plans a flash appliance dubbed Project Thunder, for later this year. The thunder-and-lightning duo takes aim at Fusion-IO, the hot-selling PCIe flash pioneer which has definitely caught EMC’s attention.


The promise of flash — or solid-state — storage is its fast response times. That’s particularly important for performance-sensitive enterprise work loads, like SAP ERP and Oracle database applications. Putting that fast flash storage on the servers running those applications, as opposed to down the wire from them, can be a huge performance boost for them.


For Project Thunder “we’re bringing together several VFCaches in a single appliance attached to the server to create even better response times,” said Barry Ader, senior director of product marketing and management for EMC’s Flash business unit. EMC’s not-so-secret sauce is its FAST software which acts as a traffic cop, directing where different data should be stored in a tiered-storage architecture. It puts the most time-sensitive data on server-based flash storage while colder data goes to more distant disks or even tape drives.


Project Thunder, which will be demonstrated at a Monday event headlined by Pat Gelsinger, president of EMC’s Information and Infrastructure Products (pictured right), shows just how much EMC is feeling the incursions of Fusion-IO and other flash players.


Critics say EMC, the storage leader, is vulnerable to nimble flash-oriented startups, given EMC’s huge installed base of non-flash storage and its big, expensive enterprise-focused sales force. Ader said VFCache will be very price competitive with Fusion-IO but would not provide pricing.


EMC, never a company to shy away from confrontation, is taking the fight to its rivals — Ader said it shipped 24 petabytes of flash storage in 2011, more than anyone. He said EMC will bring more of its tools to bear as well … adding more intelligence to its hardware and software including its acquired deduplication technology with Data Domain in 2009.


EMC pitches value of mixed storage vs. all-flash-all-the time


EMC’s view is that while webscale workloads — where Fusion-IO has taken off — are important there’s also a huge market in legacy enterprise applications that could use server-side flash for the hottest data but route less time-sensitive data to cheaper storage options companies already have deployed.


“The early adopters of PCIe technology were web scale-out early adopters. What we’re doing is bringing it to the mainstream for applications like Oracle, SQL Server and Exchange [server] — business critical applications,” Ader said.


“Many companies talk about flash but most are coming from a single [webscale] use case — that’s not what our customers are asking for. They want coordination across their systems and we’ll put the data in the most appropriate place for them,” Ader said.


Of course, Fusion-IO CEO David Flynn has a different take: What EMC is doing here “validates that flash in the server is unstoppable, but they represent the height of the mainframe era.”


One big obstacle for EMC is its cost structure — it traditionally sells into the enterprise with an aggressive, highly paid sales force. As such, EMC cannot really price VFCache as aggressively as it suggests it will, Flynn maintained. In addition, EMC sales people are used to selling to the storage people in the data center, not the server buyers. Server-side flash requires a different sale. ”The server admins hate the storage guys,” Flynn said.


It’s clear the flash storage battle has been joined. What remains to be seen is if EMC’s mixed-storage approach will prevail in a world where more workloads are flowing to webscale data centers.


Feature photo courtesy of Flickr user Lars Kasper


Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.








Friday, September 23, 2011

EMC Puts Together 1,000-Node Hadoop Care Package

Cloud Computing: EMC Puts Together 1,000-Node Hadoop Care Package — EMC, well, at least its Greenplum unit has put together a thousand-node multi-tenant analytics platform to accelerate the development and testing of the disruptive but persnickety open source Apache Hadoop software. That’s 1,000-odd hardware nodes or 10,000 nodes when you count virtual machines, along with 24PT of physical storage. In a word, that’s huge. Greenplum wants to ensure that Hadoop turns into a really serious enterprise-ready Big Data tool for sifting through mounds of unstructured data to unlock their secrets and make predictions but contends that nobody really knows how to deploy the stuff in part because everybody’s Hadoop is different. So it wants to come up with a formula others can follow and overturn the long-standing Hadoop dogma that one can’t have separate compute and storage on the same nodes. Greenplum says it increases efficiency and requires less hardware, pointing to Yahoo’s 40,000 Hadoop servers, which only offer 10%-15% utility. It thinks it can get 80%.