Cookie policy: This site uses cookies (small files stored on your computer) to simplify and improve your experience of this website. Cookies are small text files stored on the device you are using to access this website. For more information on how we use and manage cookies please take a look at our privacy and cookie policies. Some parts of the site may not work properly if you choose not to accept cookies.

sections

Amazon CTO: developers should address cloud downtime

  • Print
  • Share
  • Comment
  • Save

Application uptime in the cloud should be addressed by the developer as much as infrastructure uptime is addressed by the cloud provider. And many cloud customers get it, according to Werner Vogels, CTO at Amazon.com.

Amazon Web Services has been pushing its cloud customers to build applications across multiple of its "availability zones" for fault tolerance for quite some time now. The company started driving the point more aggressively after a major outage recently took down many websites, including some high-profile ones, like Reddit.com, Quora.com, GroupMe.com and Foursquare.

In a "post-mortem" it released following the three-day-long incident in April, Amazon said it was going to increase focus on making it easier for customers to build applications across multiple availability zones to prevent complete service outages in cases of single-site failures.

"In this event, some customers were seriously impacted, and yet others had resources that were impacted but saw nearly no impact on their applications," the AWS post-mortem statement read.

In his keynote at this week’s Structure conference in San Francisco, Vogels said users were getting smarter about building applications deployed in the cloud to be more fault tolerant. "Young businesses as well as enterprises are building increasingly sophisticated applications in the cloud," he said.

One such company is Advanced Innovations, which deployed its applications in the Amazon cloud’s availability zones on the US east coast and in Europe. "If ever the complete east coast of the US would disappear from the map, their customers can still be served out of the EU region," Vogels said about Advanced Innovations’ architecture.

Advanced Innovations designs and operates global supply-chain networks for design, manufacturing and delivery of electronic products.

Another organization Vogels used as an example of a sophisticated AWS user was the US government with its Recovery.gov website, which runs entirely on top of the Amazon cloud. "Recovery.gov is not just a simple website," he said. "It’s actually built out of highly-highly complex components."

Finally Amazon itself is a huge AWS customer. The company’s online retail outlet Amazon.com has been served out of AWS 100% since November 2010.

Vogels continued the tradition of updating the audience at the annual Structure event on growth of the AWS infrastructure. The number of objects stored in the Amazon cloud today is approaching 350bn objects. For comparison, the company was storing between 100bn and 120bn objects in the first quarter of 2010.

The company’s priorities for 2011 include adding more geographies to its global data center footprint, making it easier to build and manage applications on AWS, releasing new database offerings and broadening its support offerings.

Have your say

Please view our terms and conditions before submitting your comment.

required
required
required
required
required
  • Print
  • Share
  • Comment
  • Save

Webinars

  • 5 Reasons Why DCIM Has Failed

    Wed, 15 Jul 2015 09:00:00

    Historically, DCIM systems have over-promised and under-delivered. Vendors have supplied complex and costly solutions which fail to address real business drivers and goals. Yet the rewards can be vast and go well beyond better-informed decision-making, to facilitate continuous improvement and cost savings across the infrastructure. How can vendors, customers and the industry as a whole take a better approach? Find out on our webinar on Wednesday 15 July.

  • Is Your Data Center Network Adapting To Constant Change? (APAC)

    Wed, 24 Jun 2015 05:00:00

    Over the next three years, global IP data center traffic is forecast to grow 23 percent—and 75 percent of that growth is expected to be internal*. In a constantly changing environment and as planners seek to control costs by maximizing floor space, choosing the right cabling architectures is now critical. Is your structured cabling system ready to meet the challenge? Join Anixter's Technical Services Director, Andrew Flint and DatacenterDynamics CTO Stephen Worn and Jonathan Jew, Editor ASI as they discuss how to: •Create network stability and flexibility •Future-ready cabling topology •Make the right media selection •Anticipate and plan for density demands Essential viewing for data center planners and operators everywhere – Register Now! Please note that these presentations will only be delivered in English. 1.EMEA: Tuesday 23 June, 3 p.m BST 2.Americas: Tuesday 23 June, 1 p.m CST 3.APAC: Wednesday 24 June, 1 p.m SGT APAC customers – please note the equivalent country times: India: 10:30am; Indonesia, Thailand: 12 noon; Singapore, Malaysia, Philippines, China, Taiwan, Hong Kong: 1pm; Australia (Sydney): 3pm ; New Zealand: 5pm.

  • Is Your Data Center Network Adapting To Constant Change? (Americas)

    Tue, 23 Jun 2015 18:00:00

    Over the next three years, global IP data center traffic is forecast to grow 23 percent—and 75 percent of that growth is expected to be internal. In a constantly changing environment and as planners seek to control costs by maximizing floor space, choosing the right cabling architectures is now critical. Is your structured cabling system ready to meet the challenge? Join Anixter's Technical Services Director, Andrew Flint and DatacenterDynamics CTO Stephen Worn and Jonathan Jew, Editor ASI as they discuss how to: - Create network flexibility - Future-ready cabling technology - Make the right media selection - Anticipate and plan for density demands Essential viewing for data center planners and operators everywhere - Register Now! Please note that these presentations will only be delivered in English. 1. EMEA: Tuesday 23 June, 3 p.m BST 2. Americas: Tuesday 23 June, 1 p.m CST 3. APAC: Wednesday 24 June, 1 p.m SGT APAC customers – please note the equivalent country times: India: 10:30am; Indonesia, Thailand: 12 noon; Singapore, Malaysia, Philippines, China, Taiwan, Hong Kong: 1pm; Australia (Sydney): 3pm ; New Zealand: 5pm.

  • Is Your Data Center Network Adapting To Constant Change? (EMEA)

    Tue, 23 Jun 2015 14:00:00

    Over the next three years, global IP data center traffic is forecast to grow 23 percent – and 75 percent of that growth is expected to be internal. In a constantly changing environment and as planners seek to control costs by maximizing floor space, choosing the right cabling architectures is now critical. Is your structured cabling system ready to meet the challenge? Join Anixter's Technical Services Director, Andrew Flint and DatacenterDynamics CTO Stephen Worn and Jonathan Jew, Editor ASI as they discuss how to: • Create network stability and flexibility • Future-ready cabling topology • Make the right media selection • Anticipate and plan for density demands Essential viewing for data center planners and operators everywhere – Register Now! 1. EMEA: Tuesday 23 June, 3 p.m BST 2. Americas: Tuesday 23 June, 1 p.m CST 3. APAC: Wednesday 24 June, 1 p.m SGT APAC customers – please note the equivalent country times: India: 10:30am; Indonesia, Thailand: 12 noon; Singapore, Malaysia, Philippines, China, Taiwan, Hong Kong: 1pm; Australia (Sydney): 3pm ; New Zealand: 5pm.

  • Do Industry Standards Hold Back Data Centre Innovation?

    Thu, 11 Jun 2015 14:00:00

    Upgrading legacy data centres to handle ever-increasing social media, mobile, big data and Cloud workloads requires significant investment. Yet over 70% of managers are being asked to deliver future-ready infrastructure with reduced budgets. But what if you could square the circle: optimise your centre’s design beyond industry standards by incorporating the latest innovations, while achieving a significant increase in efficiency and still maintaining the required availability?

More link