October 30th, 2020            Click here to download the PDF file

Meeting the New Challenges of Distributed Storage Systems
-- Part 1 of 2

With the exponential growth of global data in the big data era, developing a way to safely and effectively access, store and use huge volumes of fast-moving data has become an urgent issue. As the result, distributed storage systems are receiving renewed attention and undergoing major, innovative changes. A key driver: when non-scalable systems are used for large data workloads, the systems tend to become divided and interleaved, making use of data-centric processing impossible. In order to meet the demands of the big data era and provide solutions for data-centric applications, new types of more scalable, flexible and manageable distributed storage systems are beginning to emerge.

Loxoll is now developing just such a system. While the details are not yet public, the system will employ a horizontally scalable topology, giving it the ability to serve as a replacement for traditional storage or as a private cloud solution.

To help our readers understand these problems, this newsletter highlights the key issues and challenges that the current distributed storage is now facing and what the next generation of distributed storage will likely bring.

Application of Software-Defined-Storage in Distributed Storage

These systems access the hard disk space on each network-connected servers using specialized software and integrates these scattered storage resources to form a virtual storage device. Data is distributed and then stored throughout the enterprise for effective resource utilization and management. As compared to traditional hierarchical storage systems, distributed storage vastly improves data access efficiency.

To coordinate the use of these distributed storage resources, they are organized and accessed as virtual storage devices. As such, they are managed not through physical, but virtual connection as Software-Defined-Storage (SDS). SDS orchestrates built-in storage controller functions such as volume management, RAID, snapshots, data replication and protection, etc., and abstracts them in a software layer. As such, SDS essentially serves as a storage management program. As software, SDS can run on a general-purpose server, eliminating the dependencies and complexities of proprietary software and hardware, and eliminates specific vendor dependency.

Compared with traditional storage, distributed storage offers the following advantages:

1)High adaptability. As mentioned above, SDS de-couples the dependencies of storage management functions from proprietary hardware. Customers have the freedom to choose suitable brands of hardware to run storage services on, can also use any commercial server to build a SDS-based storage infrastructure. Therefore, existing hardware can be fully utilized to meet growing storage requirements.

2)High fault-tolerance. The traditional storage hardware helps contribute to maintaining the overall system reliability. However, because this dedicated hardware no longer exists in the general-purpose server based distributed storage, the responsibility now falls on software to provide stronger fault-tolerance capabilities. SDS adopts a horizontally expandable distributed structure to minimize the instabilities introduced by a single-node on general-purpose hardware. The higher number of nodes there are, the higher the fault tolerance level will be.

3)High scalability. Traditional storage area networks are limited by the number of available nodes (devices with assigned IP addresses). SDS has no such limitation, it is highly scalable and up to unlimited expansion.

Through SDS, customers can quickly and efficiently deploy a set of highly scalable systems when implementing distributed storage solutions.


Based on the way data is formatted, organized and presented, there are three types of data storage architecture used by distributed storage: block, file, and object storage. there are significant differences in their respective capabilities as well as limitations. A distributed storage system can be designed with flexibility, taking deployment scenarios and the applicability of these methods into considerations, finding the best combination of methods so as to maximize services provided to the applications.

In addition, there are some commonly used open-source software used for distributed storage, which can assist in a more effective distributed architecture.

In the next issue of the newsletter, we will highlight the advantages and disadvantages of each when planning for a more complete distributed storage system.

For more details, your Loxoll representative can provide more information under confidential agreements in the coming months. Please contact us to schedule a private presentation.


Wanna get the latest news? Follow us on Facebook and subscribe our YouTube channel!

https://www.facebook.com/AccusysStorage/                   https://www.youtube.com/user/AccusysExaSAN