Business News

The break shows how difficult Amazon’s complex cloud backup plans are for Reuters

[ad_1]

© Reuters. FILE PHOTO: Amazon.com Inc. Participants in this year’s cloud computing conference will be following the Amazon Web Services logo in Las Vegas, Nevada, USA on November 30, 2017. REUTERS / Salvador Rodriguez / File Photo / File Photo

By Stephen Nellis

(Reuters) – Amazon.com (NASDAQ:) major companies using data services have learned a painful lesson this week, backing up the complexity of the company’s cloud drive and backing up market dominance data from other vendors, analysts and experts.

Amazon said its Amazon Web Services (AWS) data center at the Virginia data center had a “long breakdown of some network devices” on Tuesday. Interruption temporarily disrupts Netflix Inc. (NASDAQ 🙂 and Disney + commercial application playback platforms Robinhood markets Inc. (NASDAQ 🙂 as well as Amazon’s e-commerce site, which uses a lot of AWS.

An Amazon spokesman told Reuters that problems had been resolved on Wednesday.

AWS highlighted the large traces of damage caused by a single-region network problem called “US-EAST-1” that highlighted how difficult it is for companies to deploy cloud computing.

With 24.1% of the total market, according to research firm IDC, Amazon is the largest cloud computing company in the world. Like rivals Microsoft Corp. (NASDAQ :), Alphabet Google Inc. (NASDAQ 🙂 and Oracle Corp. (NYSE 🙂 AWS is trying to entice customers to use parts of their cloud, often as protection.

But working in a complex online service that can be easily switched from one provider to another in an emergency is not easy, said a senior analyst at research firm Naveen Chhabra Forrester. Instead of being a special “cloud,” AWS is made up of hundreds of different services, from basic building blocks like computing power and storage to advanced services like high-speed databases and artificial intelligence training.

Any website, Chhabra said, can use dozens of these individual services, and each of them needs to work for the site to work. It is difficult to back up to another cloud provider because some services are owned by AWS and operate very differently on another provider.

“It’s like saying, ‘Can I put an SUV body in a chassis in Berlin?’ Maybe it’s all the same and it’s on the line, but there’s no guarantee, “Chhabra said.

Another problem that makes it difficult for companies to diversify is that AWS makes it relatively inexpensive to send data to its cloud, but then charges higher prices for “exit rates” to take data out of its cloud to an enemy.

“This is exacerbated when such problems (disruptions) occur,” said Matthew Prince, CEO of Cloudflare (NYSE 🙂 Inc., an Internet security firm. I think that would increase the trust that customers have in the cloud. “

DEPENDENCIES IN A REGION

AWS itself has critical “dependencies” within its services where they link together when another fails as another may fail, said Angelique Medina, head of product marketing at Cisco Systems Inc. (NASDAQ:) ThousandEyes. That’s because AWS’s complex services are built on top of its core services. A problem that arises with a basic function such as networking can be skipped through its dependent services.

Earlier in the day on Tuesday, AWS said the disruption “affected some of our monitoring and incident response tools, which is delaying the ability to provide updates.”

Medina said AWS also appears to have gathered critical services in its US-EAST-1 region, where another disruption last year also had a major impact.

“That’s where a lot of their critical dependencies have historically been located,” Medina said. “Over time, they’ve diversified a bit.”

Chhabra, an analyst at Forrester, said Amazon has done “heavy work” to make its services resilient. But what Amazon doesn’t do for its customers is build apps so they can deal with a breakdown by hitting multiple locations or providers.

Doing so can often be an additional task that may not always be worthwhile when cloud interruptions are relatively rare.

“This is a decentralized exchange between something that is decentralized, something that is secure, and something that is available,” said Charly Fei, Interchain Foundation’s Inter Blockchain Communication product manager, who focuses on computer decentralization technologies. “It’s not something you’ll get a perfect solution for all three.”

(This story corrects typographical errors in paragraph 16)

[ad_2]

Source link

Related Articles

Back to top button