We all store (and often share) our information on a cloud. Cloud storage is typically free for limited amounts of memory (which may mean 25 GB, which is by no means a small amount...). When we assign our data to a cloud, we implicitly trust it and assume that it can provide a failsafe service, where we can retrieve all our data any time we want to. Unfortunately, that's not the case: clouds suffer from failures as our personal storage means do. And in some cases network failures add to problem, since we cannot access the cloud in the first place. Continue reading
Last week I attended the 2014 edition of the 11th International Conference on Economics of Grids, Clouds, Systems, and Services (GECON). The conference took place in the nice town of Cardiff, hosted by prof. Omer Rana of the School of Computer Science and Informatics at Cardiff University. As in the latest editions, the conference boasted a trio of chairmen: Jörn Altmann, Kurt Vanmechelen, and Omer Rana himself.
The conference is one of the few devoted to economics of ICT, Continue reading
When subscribing to a cloud infrastructure (I'm mainly interested in cloud storage, but the issue concerns computing applications no less than storage), we expect availability performance not worse than what our in-house infrastructure would provide us with. However, what do we know about the availability of public cloud platforms? The Amazon policy, for example, states that they will "use commercially reasonable efforts to make Amazon EC2 and Amazon EBS each available with a Monthly Uptime Percentage of at least 99.95%". Though this is not a strictly guaranteed Service Level, taking it as good would mean that we can expect a bit better than a 3 nines availability.
Though this is not a stellar performance, news of long outages, violating our expectations by far, appear from time to time (see, e.g., the "Another Amazon Outage Exposes the Cloud's Dark Lining" piece on Bloomberg Business Week). Many of these infos on outages appearing on the general press as well as on specialized websites have been collected and analyzed in the papers "Downtime statistics of current cloud solutions" and "The availability of cloud-based services: is it living up to its promise?", highlighting the possibility of availability performance much worse than the advertised one (if any).
Those studies are unfortunately based on reported outages, so they cannot provide a very accurate statistical representation of the phenomenon. A recently appeared paper "The Need for End-to-End Evaluation of Cloud Availability", presented at the PAM 2014 conference, provided some interesting results based on active measurements. The paper compares two approaches, based respectively on ICMP echoes (measurements carried out at the network level) or on HTTP requests (measurements carried out on an end-to-end basis). It turns out that the HTTP-method is more accurate, especially for cloud storage, where it allows to take into account the actual availability of the whole back-end infrastructure.
However, the paper falls short of giving an accurate statistical representation of the outage phenomenon in clouds. In fact, the granularity chosen by the authors is too rough for that purpose, being in the order of 10 minutes, which overlooks out-of-service glitches (most outages are indeed of very short duration). In addition, the measurements have to be really cleaned up of the outages that may be due to whatever lies in between the user and the cloud, so as to attribute clouds just the outages that are really due to the cloud provider infrastructure.
While attending the PODC conference in Paris, I had an interesting phone conversation with M., a long-time dear friend of mine (Though she will probably never read these lines, I prefer sticking to the initial of her name, just in case...). I report the content of that conversation, because it is a typical example of the misconceptions about our work. Continue reading
Last week I was in London to attend a seminar by Viktor Mayer-Schönberger at the Data Science Institute of the Imperial College. Viktor is the Professor of Internet Governance and Regulation at the University of Oxford. His research focuses on the role of information in a networked economy. His talk was on "Big Data: A Revolution That Will Transform How We Live, Work, and Think", from the title of his latest book, shortlisted for the Financial Times and Goldman Sachs Business Book of the Year Award. Though not directly involved in the mainstream of big data processing algorithms and techniques, he gave a very interesting talk on what characterizes the Big Data phenomenon and how it can be propelled by suitable data management policies. I drew a sketchnote summarizing the talk, which you can download HERE.
Through the always interesting blog of Arthur Charpentier I've seen a fascinating graphical view of many facets of the Internet. Several issues are addressed through 40 maps (you can explore the whole set of maps here), which are divided into 6 groups:
- How the Internet was created (maps 1-8)
- Internet around the world (maps 9-16)
- Threats to the Internet (maps 17-22)
- The geography of online services (maps 23-29)
- How America gets online (maps 30-35)
- How we use the Internet (maps 36-40)
It really deserves a visit!
This was the title of the keynote speech delivered by Peter Schaar at the Annual Privacy Forum 2014 held in Athens. The speaker is presently the Chairman of the European Academy for Freedom of Information and Data Protection (EAID), after having been the Federal Commissioner for Data Protection and Freedom of Information for ten years, retiring in December 2013. An appreciated, well known, and frank defensor of citizens' fundamental rights (see his portrait on the Deutsche Welle website), Peter Schaar Continue reading
Last week I was in Athens to attend the Annual Privacy Forum, jointly organized by a mixed set of organizations: ENISA (the European Union Agency for Network and Information Security), the European Commission (Directorate General for Communications Networks, Content and Technology), and the University of Piraeus (the Systems Security Laboratory).
The ever-present concerns for privacy are further heightened by the growing diffusion of big data. Nearly everyone among us has its personal data stored in some (most probably many) databases around the world, with virtually no control over them. The (lack of) actual privacy guarantees is the subject of a recent paper by Heffetz and Ligett Continue reading