On Friday August 16^th Google services incuding  Gmail, Drive and Calendar, stopped working for 2 minutes. Though only for a short time, many users noticed Google’s failure, as well as the inability to check the services’ status because Google publishes them on google.com/appsstatus which is hosted on the same infrastructure that went down.

Date	Service	Duration	Critical Data Lost
2013-08-16	Google	2 minutes	no

Reference:

http://www.zdnet.com/google-services-go-down-internet-panics-7000019516/

Permanent Link

Press Release – DRAFT – DO NOT DISTRIBUTE

Paris, June 18, 2012. The International Working Group on Cloud Computing Resiliency (IWGCR) formed in March 2012 by Telecom ParisTech and Paris 13 University has published on June 18, 2012, the first Availability Ranking of World Cloud Computing (ARWC). Average unavailability of Cloud services is estimated to 10 hours per year or more. Average availability is estimated to 99.9% or less.

Laurent Séguin, Coordinator of IWGCR, says “The Availability Ranking of World Cloud Computing provides an initial estimate of the average availability of of Cloud Computing services.”

Yingjie XU, PhD student at Paris 13 University, adds “The average unavailability of  public Cloud services can be estimated to 10 hours per year or more.” Christophe Cérin, Prof. at Paris 13 University, notices that “this estimate could be compared to the 15 minutes of average unavailability of domestic electricity in a city like Paris.”

Maurice Gagnaire, Professor at Telecom ParisTech says “Data collected for Availability Ranking of World Cloud Computing will be used to analyze the risks of Cloud Computing from a resiliency point of view with a scientific approach taking its roots in Telecommunications.”

Christophe Cérin, Prof. at Paris 13 University, Coordinator of IWGCR, concludes “The Availability Ranking of World Cloud Computing – as any ranking – is tightly related to the estimation method which was adopted. Current ranking relies on public Press reports. Cloud incidents which were not reported by Press were not taken into account. The number of users hit by an outage was also not taken into account due to the lack of public data. We encourage reading the report to understand the limitations of the current ranking and how we plan to improve it in the future.”

The report can be downloaded here: Downtime Statistics of Current Cloud Solutions (PDF). Current ranking (2007-2012) is:

Rank	Service	Yearly Unavailability (hour)	Availability(%)
1	Amadeus	0.167	99.998%
2	Facebook	0.500	99.994%
3	ServerBeach	0.667	99.992%
4	Paypal	0.833	99.990%
5	Google	0.833	99.990%
6	Yahoo!	1.000	99.989%
7	Twitter	1.167	99.987%
8	Amazon	4.000	99.954%
9	Microsoft	5.167	99.941%
10	Hostway	12.000	99.863%
11	BlackBerry	12.000	99.863%
12	NaviSite	28.000	99.680%
13	OVH	28.333	99.677%

https://iwgcr.wordpress.com/2012/03/20/international-working-group-formed-to-study-cloud-computing-resiliency

Contact Information

Name: Jean-Paul SMETS (French, English, Spanish, Japanese)
Email: contact@iwgcr.org
Phone: +33 629 02 44 25

About IWGCR

The International Working Group on Cloud Computing Resiliency (IWGCR) is a working group with a mission to monitor and analyze Cloud Computing resiliency. The working group is composed of I.T. Executives, academic researchers and industry representatives.

• Laurent Séguin (VIFIB, France)
• Prof. Maurice GAGNAIRE (Telecom ParisTech, France)
• Felipe Diaz (Telecom ParisTech, Colombia)
• Camille Coti (Galileo Institute, France)
• Christophe Cérin (Galileo Institute, France)
• Kazuhiko SHIOZAKI (Nexedi, Japan)
• Yingjie XU (Nexedi, China)
• Dr. Pierre DELORT (ANDSI, France)
• Dr. Jean-Paul SMETS (VIFIB, France)
• Jonathan LE LOUS (Alter Way, France)
• Stephen LUBIARZ (Institut Mines-Telecom, France)
• Pierrick LECLERC (Institut Mines-Telecom, France)

We are pleased to announce the lauch of our official website: http://iwgcr.org

OVH is a privately owned web hosting service company in France that provides dedicated servers, mutual hosting, domain names and VOIP telephony services. Since 2009 the company is classed as the 6th largest host of servers on the internet and the second largest in Europe

According to OVH.com, three cards of two main routers went down during 2 hours. The OVH staff will apply a corrective patch to avoid such problems in future. After that upgrade, routers must be rebooted.

Date	Service	Duration	Critical Data Lost
2012-03-27	OVH	2 hours	No

The first meeting of the International Working Group on Cloud Resiliency:

13h30 – March 23, 2012

Telecom ParisTech, 3ème étage, Salle I-304
23 avenue d’Italie, Métro “Place d’Italie”

Permanent Link

Press Release

Paris, March 25, 2012. Telecom ParisTech and Paris 13 University have formed on March 23 the International Working Group on Cloud Computing Resiliency (IWGCR) with a mission to monitor and analyze Cloud Computing resiliency. The working group is composed of I.T. Executives, academic researchers and industry representatives.

Maurice Gagnaire, Professor at Telecom ParisTech says “There is a lack of user awareness about resiliency of Cloud Computing. The main goal of our working group is to gather data about Cloud Computing availability and analyze the risks of Cloud Computing from a resiliency point of view with a scientific approach taking its roots in Telecommunications.”

Camille Coti, Researcher at Galileo Institute, says “Research in distributed computing requires realistic data to model the availability of Cloud Computing services and design efficient algorithms for resiliency.”

Yingjie XU, researcher at Nexedi, says “IWGCR will also monitor risk factors for Cloud Computing resiliency, such as misconfiguration of network routes which can lead to service outage for certain users.”. “The impact of Natural disasters such as earthquakes will also be analyzed” adds Kazuhiko Shiozaki, researcher at Nexedi.

Pierre Delort, President of the French Association of CIO (ANDSI), says “Taking data informed decisions with Cloud Computing services relies mainly on reliable sources about service availability.”

Jonathan Le Lous, Head of Innovation at Alterway, confirms that “system integrators also need trustable figures for Cloud Computing service availability in order to design hybrid IT solutions partly based on Cloud which can match their customer expectations.”

Stephen Lubiarz, manager of international relations at Mines Telecom Institute, concludes “The primary focus of the International Working Group on Cloud Computing Resiliency (IWGCR) is to provide reliable data about Cloud Computing service availability required for IT governance and IT research.”

https://iwgcr.wordpress.com/2012/03/20/international-working-group-formed-to-study-cloud-computing-resiliency

Contact Information

Name: Jean-Paul SMETS (French, English, Spanish, Japanese)
Email: contact@iwgcr.org
Phone: +33 629 02 44 25

About IWGCR

The International Working Group on Cloud Computing Resiliency (IWGCR) is a working group with a mission to monitor and analyze Cloud Computing resiliency. The working group is composed of I.T. Executives, academic researchers and industry representatives.

• Prof. Maurice GAGNAIRE (Telecom ParisTech, France)
• Felipe Diaz (Telecom ParisTech, Colombia)
• Camille Coti (Galileo Institute, France)
• Christophe Cérin (Galileo Institute, France)
• Kazuhiko SHIOZAKI (Nexedi, Japan)
• Yingjie XU (Nexedi, China)
• Dr. Pierre DELORT (ANDSI, France)
• Dr. Jean-Paul SMETS (VIFIB, France)
• Jonathan LE LOUS (Alter Way, France)
• Stephen LUBIARZ (Institut Mines-Telecom, France)
• Pierrick LECLERC (Institut Mines-Telecom, France)

The table bellow list the costs of downtime per hour:

Service	Costs
Brokerage operations	$6,450,000
Credit card authorization	$2,600,000
Ebay	$225,000
Amazon.com	$180,000
Package shipping services	$150,000
Home shopping channel	$113,000
Catalog sales center	$90,000
Airline reservation center	$89,000
Cellular service activation	$41,000
On-line network fees	$25,000
ATM service fees	$14,000

Sources: InternetWeek 4/3/2000 + Fibre Channel: A Comprehensive Introduction, R. Kembel 2000, p.8. ”based on a survey done by Contingency Planning Research.”

ZDnet reports that Facebook experienced a major downtime in Europe, Middle East and Africa. The issue probably concern DNS server according to Yfrog.

TunisiaLive reports the same issue.

The message displayed is:

Network Error (dns_server_failure)

Your request could not be processed because an error occurred contacting the DNS Server. The DNS server may be temporarily unavailable, or there could be a network problem.

Date	Service	Duration	Critical Data Lost
2012-03-07	Facebook	3 hours	No

Windows Azure service management system has been down for about seven hours worldwide. Some services were down as we can see on the Microsoft Azure dashboard. In facts, some hosted services could be hit by capacity problems.

TheRegister reports this issue on his website and Microsoft said:

We are experiencing an issue with Windows Azure service management. Customers will not be able to carry out service management operations. We have started a gradual rollout of the hotfix in North Central US sub-region. As we proceed through the rollout, we will progressively enable service management back for customers. Further updates will be published to keep you apprised of the situation. We apologise for any inconvenience this causes our customers.

Date	Service	Duration	Critical Data Lost
2012-03-01	Microsoft Azure	8 hours	No

On March 28, 2011, a 75-year-old Georgian woman shut down the Internet in neighboring Armenia for more than 12 hours. In facts, she was looking for copper near the village of Ksani and damaged fiber optic cable owned by Georgian Railway Telecom company.

According to News.az, almost all Armenia was left without internet and some areas of Azerbaijan were also affected. On Bloomberg.com, we learn that Georgian Railway Telecom LLC spokesman Giorgi Ionatamishvili said that “the cable belonged to the company and customers suffered massive and catastrophic damage. The information is available on PCMag website or Theguardian.

A spokeswoman for the country’s largest Internet service-provider Armentel, Anush Begloyan, said “the incident forces our company to think about diversifying our channels.” according to Mb.com.

Date	Country	Nature	Consequence
2011-03-28	Armenia	Accident	All of Armenia offline for about 12 hours