Skip to main content

privacy on the cloud

There is a huge variety of cloud-based data archival and backup services available today. The following were shortlisted based on certain desirable features such as encrypted storage, personal encryption (zero knowledge), storage size and bandwidth limits and cross-platform support. 


The pricing structure and other particular features are outlined below. The order reflects the slight preferences of the author.
  1. MEGA
    • # machines : ?
    • Pricing
      • 50GB – Free
      • 4TB - $12/m
  2. Crashplan
    • # machines : up to 10 computers
    • Pricing
      • Unlimited - $13.99/m, $150/y
    • Other features
      • de-dup
      • can run server-free, exchanging backup space with friends and family
  3. AltDrive
    • # machines : 3 computers
    • Pricing
      • Unlimited - $9.95/m, $99/y
    • Other features
      • de-dup
  4. Bitcasa
    • # machines :  up to  5 devices
    • Pricing
      • 20GB - Free (up to 3 devices)
      • 1TB - $10/m, $99/y (up to 5 devices)
      • 5TB - $49/m, $499/y (up to 5 devices)
      • Unlimited - $99/m, $999/y (up to 5 devices)
  5. Elephant Drive
    • # machines : up to 10 devices
    • Pricing
      • 100GB - $9.95/m, $100/y
      • 250GB - $19.95/m, $201/y
      • 1TB     - $85.95/m, $86/y
      • 2TB - $169.95/m, $1713/y
  6. Wuala
    • Pricing
      • 5GB – Free
      • 50 GB - $6.99/m, $75/y
      • 100GB - $11.99/m, $129/y
      • 200GB - $22.99/m, $249/y
      • 500GB - $54.99/m, $599/y
      • 1TB - $99.99/m, $1099/y
      • 2TB - $179.95/m, $1999/y
  7. Spider Oak
    • # machines : unlimited
    • Pricing
      • 2GB – Free
      • 100GB - $10/m, $100/y
      • 200GB - $20/m
      • 500GB - $50/m
      • 1TB - $100/m
    • Other features
      • de-dup


Diversification
A prudent cloud-backup strategy would involve using multiple services – diversified geographically, both in terms of the location of the company and the location of the data. So, for e.g., a combination of MEGA, Crashplan and AltDrive, would satisfy both criteria.

Local Encryption
Locally securing the data in an encrypted vault, using tools such as BoxCryptor,  TrueCrypt or FreeOTFE (see here for a comparison), before sending along the wire, adds an additional layer of security and privacy. "There is a potential backup trap when it is used for files that will be synced or stored in the cloud. Encryption programs that create encrypted “volumes” (files that contain encrypted files) do not change the size of the volume (container file), and often – intentionally – do not change the modified date of the volume, even though files in the volume have been changed or added. The result can be that your cloud service does not recognize that the volume file has changed, and will fail to update the online copy. TrueCrypt is an example of an encryption program that does not change the modified date of volume files (encrypted file container). However, some cloud backup services – Dropbox for example – check the hash value of volume files, not the date, and if that changes Dropbox stores the latest copy of the volume file. If you’re using Dropbox, that makes TrueCrypt an excellent way to implement client-side encryption for your most sensitive files. SkyDrive, monitors the modified date – not a hash value – so TrueCrypt volumes are not updated in the cloud by SkyDrive after their content changes client-side."(quoted verbatim from [2])

The Bottom Line
  1. At the very least, use an online back service that provides the feature of personal encryption (in addition to encrypted storage). This means that nobody in the company can easily break into and peer at your data. This also means, however, that the data can be irretrievable if one forget their personal key/password etc. All of the services listed above provide personal encryption (except, on a technicality, Bitcasa -- but their approach is quite similar).
  2. Optionally, diversify your data and use multiple services.
  3. Optionally, vault your data locally and then backup the vault.
    1. This option can also be used in conjunction with a service that does not provide personal on-the-fly encryption

Sources 


Comments

Popular posts from this blog

the brave new economy

A free(er) market is emerging and has been emerging ever since the advent of the Internet and the Web. Newer technologies have accelerated this. Old inefficiencies in knowledge and access to the market are quickly disappearing. Participation is increasingly peer-to-peer (P2P). Participants can produce, create, curate, give, lend, sell, share, auction, consume, take, borrow, buy, reuse, rent and/or barter goods, services, cash, credit, currency, equity, debt and/or knowledge. The resultant economy is highly collaborative and is sometimes referred to as the ‘ sharing economy ’. Efficiencies are being introduced and value is being created in the smallest of markets. Non-profit activities are thriving as well, in addition to for-profit ones. The marketplace is becoming, at once, global and local. An important subset of the new economy is collaborative and crowdsourced , and may be described as commons-based peer production . Asymmetries in information are diminished by algorithms (Google

Migrating from MS SQL Server to MongoDB

The following contains notes from various attempts at migrating 2.5GB of MS SQL Server data to MongoDB, on an 8 GB, quad-core, 64-bit Windows 7 Enterprise machine. [TERMINATED]  http://mssql2mongo.codeplex.com/ : Simple to use. Exponential slowdown observed in migration. For a total of the 50 million records spanning two tables, the following migration times were observed: 1 second for the first 100,000 records,  30 minutes for 1,000,000, 20 hours for 16 million (after which I terminated the process). [DID NOT WORK] http://rubydoc.info/gems/mongify/ : A ruby-based approach. Use Ruby 1.9.3 (tiny_tds dependency causes problems with Ruby 2.0). Install DevKit before installing the mongify gem. Also, use ' sqlserver ' as the adapter in the .config file. Then, before running ' mongify check <config-file> ', install the gems  activerecord-sqlserver-adapter  and tiny_tds . At this stage, it fails. ' mongify translation <config-file> ' fails as well.

Technology companies ranked by revenue per capita

As of April 19, 2016, we have the following ranking of some prominent technology companies, in order of revenue per employee: Company Revenue Per Employee Apple $2M Netflix $1.9M Facebook $1.37M Alphabet $1.2M Microsoft $0.79M Qualcomm $0.76M Cisco $0.68M PayPal $0.58M Intel $0.51M Juniper Networks $0.48M NetApp $0.48M EMC $0.48M Yahoo $0.47M Amazon $0.46M Adobe $0.38M Salesforce $0.35M VMWare $0.33M LinkedIn $0.32M Citrix $0.31M SAP $0.31M Oracle $0.28M eBay $0.24M IBM $0.21M It is interesting to note that these numbers are far exceeded by pharmaceuticals (e.g. Gilead, $3.2 M) and simply dwarfed by energy companies (e.g. Valero, $13 M and Philips $10 M).