Google Outage Shows Public Cloud Computing Is ‘Not Invincible’

‘There are going to be outages and shutdowns for public cloud,’ says Bob Venero, CEO of Holbrook, N.Y.-based solution provider Future Tech, No. 96 on the CRN 2020 SP500. ‘Corporations that rely solely on public cloud to function and run are putting their businesses at risk for loss of revenue, security vulnerabilities and unexpected losses in productivity. Public cloud is not invincible.’

ARTICLE TITLE HERE

A widespread Google Cloud outage that took down multiple Google services, including YouTube and Google Workspace (formerly G Suite) and its Gmail service, on Monday morning is yet another sign that public cloud computing is not invincible, partners said.

“There are going to be outages and shutdowns for public cloud,” said Bob Venero, CEO of Holbrook, N.Y.-based solution provider Future Tech, No. 96 on the CRN 2020 SP500. “Corporations that rely solely on public cloud to function and run are putting their businesses at risk for loss of revenue, security vulnerabilities and unexpected losses in productivity. Public cloud is not invincible.”

In a statement, Google said at 6:47 a.m. EST, it experienced an authentication system outage for approximately 45 minutes due to an internal storage quota issue.

id
unit-1659132512259
type
Sponsored post

“Services requiring users to log in experienced high error rates during this period,” a Google spokesperson said. “The authentication system issue was resolved at (7:32 a.m. EST). All services are now restored. We apologize to everyone affected, and we will conduct a thorough follow-up review to ensure this problem cannot recur in the future.”

The Google Cloud outage follows a high-profile outage at Amazon Web Services last month that affected thousands of third-party online services and dozens of AWS services. It also comes after several Microsoft cloud outages on Sept. 28, Oct. 1 and Oct. 7.

“You can’t assume the public cloud is always going to run,” Venero said. “You really need to assume there are going to be outages. It is smart for businesses to protect themselves with a hybrid approach. Applications that are mission-critical to your organization’s growth and success must be on premises.”

Many Fortune 500 organizations are adopting an all-public cloud strategy that is “risky” to their companies and could ultimately result in a “loss of revenue and a blow to their credibility and stock price,” Venero said. “No one gets fired for moving to public cloud until there is an outage. There is always a scapegoat when there is a big outage.

Future Tech, for its part, expects outages to become more and more frequent as more companies adopt public cloud services. In fact, Future Tech urges its customers to carefully calculate the cost of outages when it adopts a public cloud solution, Venero said.

Venero echoed comments from Dell Technologies CEO Michael Dell that on-premise solutions are more cost-effective 85 percent to 90 percent of the time.

“Dell Technologies has done a great job putting together great solutions that protect businesses with a hybrid approach that bursts out to the cloud,” Venero said. “A lot of businesses move to the public cloud because it is easy, and there is no capital expenditure. But, in the long term, it costs them more money. That has been proven time and time again. We had one large company that had an $11 million charge over-run in a single quarter because of egress fees associated with moving data back and forth for a development environment.”

The Google outage was first posted on the Google Apps status dashboard at 6:55 a.m. EST, impacting Google Gmail and all editions of Workspace, YouTube and even the Google Nest security service, said Allen Falcon, founder and CEO of Cumulus Global, a 14-year-old Westborough, Mass.-based cloud solution provider. He said it appeared YouTube, Nest, Workspace, Google Voice and Google Analytics were back up at 7:52 a.m. EST.

Falcon suspects 100 percent of his Google Cloud customers were affected by the outage but, because it started before 7 a.m., only 10 customers reported the problem to Cumulus Global. “It was a pretty quick response by Google to get it back up and running,” he said. “I’ll have to crunch the numbers to see if this puts customers over Google’s promised 99.9 percent SLA (service level agreement).”

Given the full range of Google services that went down, Falcon suspects the cause was an account authorization issue. “It looks like there was an inability to connect to services because of an account authorization issue with one Google account for all these services,” he said. “What is common across all these services impacted is the authorization account management structure. We are still trying to confirm that.”

The Google Workspace dashboard reported at 8:31 a.m. EST that Google Cloud was still investigating reports of problems with Gmail. “The affected users are able to access Gmail, but are seeing error messages, high latency and/or other unexpected behavior,” a dashboard post said.

At 8:56 a.m., Google Cloud reported that “the problem with Gmail should be resolved. “We apologize for the inconvenience and thank you for your patience and continued support,” the dashboard post said. “Please rest assured that system reliability is a top priority at Google, and we are making continuous improvements to make our systems better.”

Hybrid-cloud and multicloud make sense for midsize and enterprise customers, but most small businesses do not have the resources to run a hybrid cloud environment or to manage multiple clouds, according to Falcon.

“For small businesses, public cloud is still more secure, reliable and cost-effective than on-premise,” he said. “Large organizations can afford backups and generator systems to remain online. Most small businesses do not have the money to spend for battery backups and generators or multiple internet connections. That is something that is typical in midmarket and enterprise customers. The economics are different in SMBs because of resource availability.”

Public cloud usage is soaring in the wake of the global pandemic, which could be causing stress on public cloud, Falcon said. “Public cloud utilization is way up, so it is possible these cloud services may be hitting some of their capacity constraints,” he said.