Data Center Tech Blog

community moduleOpen source software is not new in the enterprise datacenter. In fact, Linux and other open source tools and utilities have been growing steadily over the past few decades, and are used in most datacenters. Over the past 20 years, open source has matured with increased participation, commercial backing, mature business models and most of all broad awareness and acceptance. With commercial companies such as Red Hat, SUSE and others backing Linux and providing support and services, it has become a trusted staple for mission critical applications. There is now room being made in storage systems for open source based solutions.

Attending OpenStack Forum in Atlanta 2 weeks ago showed me how open source developed software and functionality has fast become a go-to source for new cloud and hyperscale architectures as well. This conference has grown from 1800 people in San Diego just 18 months ago to 4500 in Atlanta this month! There are over 200 large corporations involved and 1000s of developers worldwide actively engaged in the over 100 independent projects that work with the OpenStack APIs and frameworks.

So Why Has OpenStack Grown So Fast While It Took Linux Over 20 Years?

The OpenStack movement learned from the Linux experience and has put many structures in place to accelerate its growth. No doubt, the need for speed, agility and fast scaling in the world of cloud and hyperscale had a great deal to do with why OpenStack itself has grown so rapidly. Learning from other, earlier open source undertakings, the 100’s of projects involved in OpenStack are linked through the OpenStack Foundation which provides infrastructure and governance. The broad acceptance of open source as a viable model and the proven nature of open source business models have spurred a breadth of community and commercial involvement and choices. Another important output from OpenStack has been frameworks and API for a number of projects and companies to work together and interoperate. This has been key, as working in a datacenter involves working across various resources and utilities. While the early open source community had to learn through trial and errors and better community organization, OpenStack has benefited from previous open source learnings and has put it to good use.

An Ecosystem to Rely On

openstack summit 2014

Photo by: opensource.com

The breadth of the ecosystem of vendors and projects that serve the datacenter and are a part of OpenStack is staggering. It also caters to the new application developer who wants to be able to develop apps quickly and easily and to feel empowered to do so without having to contact numerous IT professionals. The ecosystems spans virtualization solutions like KVM, Zen, VWware to compute vendors and solutions like Intel, HP, Dell, Red Hat, SUSE, Oracle, storage vendors such as EMC, NetApp and networking giants such as Cisco and Juniper. These are not small companies, nor only open source projects. These are some of today’s leading datacenter technology providers.

OpenStack provides a viable alternative to more proprietary architectures to build private, hybrid and public clouds. The Tier 1 cloud vendors like Amazon Web Services and Microsoft Azure have their own architectures. Many tier 2 cloud vendors and enterprises have been looking for architecture to adopt and the latest and 9th release of OpenStack called Icehouse is getting some serious traction.

OpenStack and a New Storage Paradigm

cloud storage problem solving

Scale-out architectures is the norm for storage in cloud-centric and hyperscale datacenters. Storage needs to scale easily and quickly while handling the volume and variety of data in both structured and unstructured form. OpenStack provides for both an object interface and a block interface with projects like Ceph, Swift and Cinder and these are fast becoming the basis of new storage architectures.

The storage and infrastructure world have been looking for more cost effective alternatives as they move to the cloud and operate at hyperscale level and OpenStack provides this. It allows for self-service DIY models and can offer far more economical solutions than what has previously existed.

From the user’s perspective, the benefits are around the pluggability and programmability of the APIs provided. Other benefits include the access to a large ecosystem of projects and companies, and the support for new and custom workloads such as media streaming and web serving. There are major telecommunications, media and financial companies who are using OpenStack and it is being rolled out to more and more datacenters.

As a flash storage vendor working with storage software stacks of all kinds, it is hard to ignore the momentum and growing maturity of OpenStack. With flash becoming an important storage fabric enabling faster performance and greener datacenters, it is clear to me that OpenStack is an important ecosystem to work with in order to serve cloud and hyperscale datacenters.

enterprise readiness

Enterprise Readiness

There is still work to be done on the management and operational side as well as the networking side. There is also a shortage of people skilled in working with OpenStack and training is needed to enable more people to work with an OpenStack infrastructure. But it is clear to me that the momentum will continue with the thousands of people, 100s of companies and projects engaged and leadership from the foundation to move things forward. The new world of cloud, social, mobile and the hyperscale and their data growth indicates that room will be made for OpenStack in storage, and that companies and the community will work together to make it enterprise ready.

Join me to talk SSDs and Open Source next week at the SV DB Meets Solid State Meetup!

Tuesday, June 3, 2014 at 6:30 PM to 9:00 PM

Next week the Database Meets Solid State, a Silicon Valley based community group that provides a meeting point for people interested in the intersection of flash memory and database acceleration, will be meeting at the SanDisk® headquarters in Milpitas to talk about what works and what doesn’t for SSDs, OpenStack, open source and cloud storage technologies.

This is going to be a great discussion with David Floyer, CTO of Wikibon, James Malachowski, CEO of NodePrime and myself! You can find out more information on each speaker and topic on the Meetup homepage.

The event will take place on Tuesday, June 3, 2014 at 6:30 PM to 9:00 PM, at SanDisk, 601 McCarthy Boulevard, Milpitas, CA.

I look forward to seeing you there!

– –

What are your experiences with OpenStack and open source technology in the storage industry? Share your thoughts with me at @NithyaRuff on Twitter, and join the conversation with @SanDiskDataCtr

SanDisk® Enterprise at IBM Edge 2014

Enabling the Flash Transformed Enterprise at HP Discover 2014

subscribe blog