The Datomia® innovation is in the merging of NAS technology and object storage. In Datomia® storage, every file is also an object, with a unique ID. It has the ability to be accessed from any application, from any geographic location and from any public or private storage provider, with simple HTTPS protocols, regardless of the same object being filed in a sub-folder on the NAS file system. This provides enterprise applications with a multi-vendor storage solution that has all benefits of object storage.
Datomia Takes the Advantages of Both Systems, and Minimizes Their Weaknesses
Traditional NAS storage offers very good performance. However, it produces too much replicated data, and it has very demanding infrastructure requirements, such as stacks of SSDs, Flash Drives and other dedicated hardware to provide performance.
Object storage is much simpler. The data is secure, and it may be dispersed (although, even with erasure coding, it will require an additional 60% data duplication for disaster recovery). However, you cannot replace a NAS with an object store. Object storage is missing the basic protocols, interfaces and hierarchical structure of NAS storage systems. This means it requires complicated integration scenarios with existing enterprise applications.
Erasure-Coded Object Storage Underneath
Datomizer® is the trademark for Datomia®’s distributed, scalable, fault-tolerant and highly available cloud vault storage product. The Datomizer® NAS appliances have erasure-coded object storage underneath, similar to the differentiated cloud storage solutions provided by AmpliData, Cleversafe and SwiftStack.
Object storage has become a viable alternative to scale-out network-attached storage (scale-out NAS), because of its unlimited scalability, lower emphasis on processing and high-capacity networks, access via Internet protocols (rather than storage commands), custom metadata and off-the-shelf component compatibility.
You Can Mix Storage Nodes from Multiple Vendors
The user has the choice of storage providers, including on-site, and they can switch between them at will. Unlike solutions offered by individual cloud providers, Datomia® lets you mix storage nodes from multiple vendors. Even if a governing body pressured a cloud storage provider to relinquish all data, the data contained within Datomizer®’s Data AtomsTM could not be reconstructed without access to the locally installed NAS Vault.
Although the vault can be retrieved and recovered from the cloud, the keys will not be available in the cloud. This is because the client maintains the keys.
The Block and File System Storages It Uses Are Both Mature and Proven Technologies
Datomia® enhances block and file system storage to meet the needs of an increasingly distributed and cloud-enabled computing ecosystem.
With block-based storage, blocks on disks are accessed via low-level storage protocols, such as SCSI commands, with little overhead and no additional abstraction layers. It's the fastest way to access data on disks. All higher-level tasks, such as multi-user access, sharing, locking and security, are usually deferred to operating systems. In other words, block-based storage takes care of all the low-level plumbing, but it depends on higher level applications for everything else.
High Performance and Secure Erasure Coding
Datomia® has created its own erasure codec for implementing very secure cloud NAS storage with a simple file system. This is the component responsible for the slicing of information across a number of storage nodes.
The structure of the erasure code is optimized to the user’s preferences
This structure provides erasure coding with flexible parameters. Users can manage storage efficiency, data protection against failures, network traffic and CPU utilization. To ensure the lowest latency, Datomia®’s erasure codec distributes network traffic, in conjunction with a special method for load balancing.
Datomia®’s codec configures an erasure correcting code from component codes of smaller lengths
Datomia® created a library of component codes, consisting of the best maximum distance separable (MDS) codes (such as Reed-Solomon), and codes with low encoding/decoding complexity (such as optimal binary linear codes).
Storage efficiency is enhanced by using MDS component codes. Network traffic and computational complexity are reduced by using linear codes over small finite fields. The number of component codes within the configured erasure correcting code depends on the number of available storage nodes; this is determined by the vault’s structure.
A High Performance, Software-Only Solution
Most other algorithms are limited in their ability to change the number of storage repositories the system can use. Datomia® is different. Its erasure codec has an improved performance algorithm for data processing. Datomia®’s codec maximizes IOPS ratio by using concurrency and parallel processing. This avoids the intensive CPU utilization and latency, which are common problems with erasure coding in many of today’s erasure coding-based solutions. As a result, Datomia® can be software only and doesn’t require a hardware device for encoding files. It also avoids operational limitations within datacenters.
Datomia® also combines smaller 4KB-128KB blocks into bigger slices, which significantly optimizes and increases overall performance. In practice, this means the Datomia® codec requires far less processing power than the dedicated nodes of a Cleversafe, and it is far easier to deploy.
Higher Levels of Security
To achieve higher levels of security (and protect customer’s data from unauthorized access and theft), the Datomia® erasure codec mixes and hides the customer’s original data. This is while operating within public or private cloud premises.
The degree of data mixture is selected according to user preferences. The mixture degree is typically the smallest number of storage nodes that need to be accessed in order to reconstruct the chosen amount of original user’s data. Higher mixture degrees correspond to higher levels of data protection against unauthorized access and higher data retrieval complexity.