Learn how HPE used Delphix technology to accomplish the largest ever corporate separation.
Share
Breakups are never easy – but what if you're trying to split a $100 billion company into two independent leaders? Who manages the IT Infrastructure?
When HP was facing a corporate separation, it needed to divide 2,800 applications, 300,000 employees, 75,000 application interfaces and their network connectivity, 570 projects and 50,000 servers1. In the company's 2014 Q4 earnings call, CEO Meg Whitman called this undertaking "a big and complicated separation -- the biggest ever done. This is not a spin-off, but two Fortune 50 companies, both with about $57 billion in revenue."
Under the leadership of CIO Scott Spradley, HPE enlisted a team of their top talent to tackle this herculean challenge - one that involved splitting 6 data centers2 - with the help of Delphix. Now, they've been named one of this year's CIO 100 winners for their successful efforts. Read how CIO summarizes the project here.
Mahesh Shah, General Manager of M&As at HPE, said of this undertaking, "Infrastructure, data and security - those were the most critical elements and the most complex elements for us to go and separate into two companies."
All three of those discrete challenges are areas in which Delphix excels. Although data virtualization had never before been used to support a corporate separation, much less one of this size, HPE leadership recognized conventional methods would not be fast enough to meet their aggressive deadline. They looked towards a relatively new technology for this massive, mission-critical project, and by the project's conclusion, the HPE team had applied Delphix software through multiple stages, including data conversion, regression testing, business continuity, data archiving and legal hold.
HPE achieved four major benefits through its engagement with Delphix:
A dramatic reduction in application downtime
Retention of a complete copy of the applications, with 70% data compression
Disaster recovery capabilities to "roll back" the applications in case of accidental data deletion
Reduced infrastructure requirements of over 50%
HPE continued to amaze watchers-on and their internal team alike. Mr. Spradley explained, "Nobody ever thought that we would do design in a month, or that we would do build in a month, or that we would write 172,000 [System Integration Test] cases, that we would test 74,000 interfaces, that we would split up 50,000 production servers in the amount of time. You typically build 900 servers in a year, maybe 1000 for an IT organization. We built 4000 servers in six weeks."
We'd like to congratulate HPE for this exciting win. This talented and hard working team was able to execute this split efficiently, in 9 months with only two hours of downtime. For a process that could have taken years, and with a minimum of sixteen costly hours of downtime, the time and cost savings are nothing short of spectacular.
We will be proud to watch Mr. Spradley accept this award on HPE's behalf at the CIO 100 Symposium & Awards Ceremony, taking place August 14-16 in Rancho Palos Verdes California. Join us for a discussion at the event, or learn more about how Delphix can help your company through similarly complex, business-critical problems, contact us at sales@delphix.com.