JaCEnt

Amazon’s Cloud Storage Solution S3 – And Beyond  

Amazon’s cloud storage solution S3 is the focus of JaCEnt’s attention this week. As part of our on-going commitment to ensure all data we store is secure and highly available, we will be reviewing the S3 and asking the question: does it meet our high standards and is it good enough for our customers?

Introducing the Amazon S3  

The Amazon S3 can be used to back-up, archive and retrieve any amount of data files or objects. With a Service Level Agreement that commits to an incredible monthly uptime of at least 99.9%, high availability certainly rates as one of Amazon S3’s best features. It has been deliberately designed with a minimal feature set for simplicity.

Most impressively, it has also been reported that Amazon could suffer catastrophic failures in two of their data centres and still avoid any detrimental effect on the data stored within S3.

Amazon’s S3 ticks the security box too – it provides several data protection services, as well as encryption for secure transfer and controlled access. Despite the impressive high availability and security of Amazon S3, there is one performance flaw. By design, the S3 storage architecture does not support incremental changes.

​Consider this very common scenario: you are accessing a single webpage. To simply write an entry to the access log, the web server you are connected to would have to pull said log from S3, write it to the local cache and then push it back. For JaCEnt, this is not an acceptable set-up. We loved the high availability of S3, but we would be ashamed to offer such read\write speeds to our customers! 

So we began the search for an alternative storage solution… 

Enter Fused-Based File Systems and S3QL  

FUSE (Filesystem in Userspace) is a loadable kernel module for Unix-like computer operating systems – it lets non-privileged users creates their own file systems without editing kernel code. This is achieved by running file system code in user space while the FUSE module provides only a “bridge” to the actual kernel interfaces.

​As a result, this allows genius developers to create tools such as S3QL, a file system that stores all its data online using storage services like S3 and effectively provides a hard disk of dynamic, infinite capacity. A perfect solution! 

With so many S3 file systems out there, why do we favour S3QL? 

S3QL has many benefits and a whole host of additional features that make it particularly useful for online storage; we’ve listed the best of these below:

Data Storage: S3QL provides the capacity for infinite data storage, which can be accessed from any computer with internet access in any location.

Encryption: S3QL provides additional security. All data can be AES encrypted with a 256 bit key after compression. Furthermore, a further SHA256 HMAC checksum protects the data against manipulation. Such effective security measures ensure that JaCEnt conforms to strict PCI guidelines, both for our corporate website and customer sites.

Data De-duplication: This feature reduces the cost of storage and allows us to pass on savings to you. If several files have identical contents, the system will store the data only once. This works across all files in the system, including those where only parts of the file are identical.

Immutable Trees: Directory trees can be made immutable, preventing their contents from being altered in any way. This is a great feature for securing backups as it prevents modification after they have been made.

Improved Bandwidth: S3QL has implemented a feature to improve bandwidth and increase the response time for websites. File contents are split into smaller blocks that are stored locally. This reduces the number of network transactions required to read and write data, making S3QL far superior to S3 in terms of read\write speeds.

We love S3QL, but what do you think? Do you agree, disagree or have a burning question? Be sure to leave your comments and come back soon for our next blog.