Cleversafe Boosts Storage Ingest Rates to 1TB/s for 10EB Storage System

Cleversafe Inc. announced a series of storage appliances, based on Intel Xeon processors, to achieve performance level throughput for ingesting and storing data at exabyte scale.

Cleversafe’s new 3000 series of appliances can be configured in a system capable of capturing data at 1TB/s at exabyte capacity, making the company’s Object-based Dispersed Storage solution effective for high levels of data throughput.

"The inherent problem that companies are struggling with is building the right infrastructure to enable the world’s data scientists to make accurate predictions. That means being able to store more data over longer periods of time," said Russ Kennedy, VP of Product Strategy, Marketing and Customer Solutions for Cleversafe. "At Petabyte and Exabyte scale the ability to ingest and store those volumes of data requires a significant boost in performance throughput," said Kennedy. "We can scale this system to achieve an ingest rate of 1 Terabyte/second – that’s 20,000,000-3MB photos per minute, 180,000,000-20MB digital x-rays per hour, and even 1,728,000-50 GB HD videos per day. To finally be able to achieve that scale – that’s a serious shift in business for any company."

Cleversafe’s new 3000 series appliances allow thousands of simultaneous readers/writers to maintain continuous, aggregate performance at a data throughput rate of 1TB/s in a 10EB system configuration. Cleversafe’s Object-based Dispersed Storage technology combined with the performance benefits of Intel Xeon processors delivers reliable, cost effective, and power efficient big data storage solution.

"Today’s business requirements for capturing, storing and analyzing data at massive scale require CIOs to optimize their infrastructure strategies. Intel Xeon processors help solve big data storage challenges by providing energy efficient performance, scalability and built-in data protection required for today’s storage systems, " said David Tuhy, Intel Corp.‘s Storage GM.

Cleversafe is eliminating limitations on the movement, management and storage of massive data stores for analytical purposes. This past year the company announced the capability to store data at the 10EB scale, and recently the company announced it was combining computation with its dispersed storage solution to deliver high reliability and efficiency to users of Hadoop MapReduce.

This announcement here continues to show how Cleversafe is shaping the new data storage and analytics landscape to solve the challenges of big data.

"Petabyte and Exabyte scale data sets simply exceed the boundaries of normal data processing capabilities. When you get to those levels of scale you need to take a non-traditional approach to storage for cost-effectively capturing, storing and analyzing data," said Terri McClure, senior analyst at Enterprise Strategy Group. "With traditional storage systems there are too many potential performance bottlenecks to capture data flow at such volumes. Cleversafe’s new series of storage appliances is a non-traditional approach that sets an extremely high ground for performance. This should help enable its customers to realize the real benefits associated with leveraging long-term Exabyte-scale data stores for analytics to deliver real business value."
 
Product Features:

• The Accesser 3100 model is powered by dual 8 core Intel Xeon ProcessorE5-2650 CPUs. It features 10GbE saturation on multiple ports to provide maximum throughput per Accesser.
• The Slicestor 3510 model features high-density drive enclosures for efficiency – 252TB to 336TB respectively (84 3TB drive enclosures in 5U with 84 4TB drive enclosures in Q4).
• The dsNet Manager 3100 model is capable of managing up to 100PB with a single appliance.

Cleversafe’s new 3000 appliance series models will be available this Fall.

Cleversafe’s dsNet system protects both data and metadata equally and is more reliable than traditional storage systems. By applying the company’s Information Dispersal technology to slice and disperse data, single points of failure are eliminated. As data is distributed evenly across all Slicestor nodes metadata can scale linearly and infinitely as new nodes are added, thus reducing any scalability bottlenecks and increasing performance and reliability. Its approach delivers the combination of analytics and storage in a geographically distributed single system allowing organizations to scale their big data environments to hundreds of petabytes and even exabytes today.