By Jane Clabby, Clabby Analytics
On November 15, 2016, Western Digital introduced a new addition to its object-based storage (OBS) solution portfolio — the ActiveScale P100. The integrated turnkey system is an easy-to-deploy entry-level system that scales modularly from 720TB to 19PB of raw capacity, and is designed for Big Data applications across on-premise and public cloud infrastructure in a range of industries including life sciences, media and entertainment and government/defense. Included in the new offering (and also available for the existing Active Archive System) is ActiveScale CM (cloud management), a new cloud-based monitoring tool that provides remote system health monitoring and predictive performance and capacity analytics.
According to IDC, file and object storage is expected to be a 32 billion dollar market by 2020. Comprised of primarily unstructured data sets, these large volumes of data are being used increasingly for Big Data analytics in applications such as fraud detection, machine learning, genomics sequencing, and seismic processing.
The Western Digital OBS family includes the new ActiveScale P100, and the Active Archive System. Both are scale-out OBS solutions that provide the benefits of object storage — including massive scale, easy and immediate access to data, and data longevity. Vertical integration makes these systems easier to buy, deploy and manage, and the tuning and optimization provide better performance than solutions that include white box or other DIY components.
Major features of the new ActiveScale P100 and existing Active Archive System include:
- Amazon S3 compliant scale-up and scale-out solution -ideal for Big Data applications.
- Strong consistency ensures that data is always up-to-date.
- BitDynamics continuous data scrubbing provides integrity verification and self-healing.
- Advanced erasure coding offers better data durability (up to 15 nines) than traditional RAID systems.
- BitSpread provides data protection without replication, so capacity requirements are reduced.
- ActiveScale SM (system management) provides a comprehensive, single-namespace view across scale-out infrastructure.
- ActiveScale CM is a new cloud-based monitoring tool that provides remote system health monitoring and predictive performance and capacity analytics for multiple namespaces.
Active Archive Customer Examples
- Ovation Data, a data management company, uses Active Archive System in conjunction with Versity Storage Manager to build private storage clouds for customers in the oil & gas industry. The company selected the Active Archive System because it provided the economics of tape storage with the performance of disk storage. The solution provides cloud-scale storage, automated tiering to object storage and tape, and speedy cost-effective access to data stored over a long period of time — improving efficiency and enabling users to make data-driven decisions that will drive business value . The system also provides a direct plug-in to VSM without any modifications for quick, easy deployment.
- EPFL (École Polytechnique Fédérale de Lausanne) is using Active Archive System for the Montreux Jazz Festival Project, a digital archive storing 50 years of the festival’s audio-visual content and accessible by academics for research and education purposes, and ultimately by the general public for their listening pleasure.
ActiveScale P100 – A Closer Look
The ActiveScale P100 is a modular turn-key system developed based on customer demand for an easy-to-deploy system at an entry-level capacity, performance and price-point.
The system includes three 1Usystem nodes combined with 10TB HelioSeal drives that form a cluster (metadata is copied on all system nodes) with a 10Gb Ethernet backplane, and provides the ability to easily snap pieces together in 6U increments to expand the system. Scale–out configuration options include a scale-out capacity configuration expandable to 9 racks and 19,440TB of raw capacity, as well as a scale-out performance configuration expandable to 2,160TB of raw capacity and throughput up to 3.8GB per second. Configurable erasure coding allows software policies to be set along a durability and capacity continuum to determine the best match for a given workload.
ActiveScale SM provides dashboard-based real-time system management and monitoring of system health to proactively identify any potential problems such as performance or capacity issues. Also managed through this comprehensive view is installation, configuration, upgrades and capacity expansion. Management wizards automate many common functions.
ActiveScale CM collects and analyzes historical system data to identify trends and patterns for operational and business insight. ActiveScale CM can correlate and aggregate multiple metrics across the entire namespace for a comprehensive view of storage infrastructure or can examine specific elements such as individual geographies or applications, for example, to identify potential problem areas —providing better management of SLA’s, improving efficiency, and reducing management costs.
As businesses are recognizing the value of collecting and analyzing years worth of data —largely unstructured data collected from a wide variety of sources—traditional storage options don’t provide the flexibility or scalability required to support today’s Big Data applications. Tape can cost-effectively store data for many years, but isn’t easily accessible. SAN and NAS storage weren’t designed to support cloud-scale applications or to store unstructured data such as images or video. So many businesses have turned to object storage to overcome these limitations. Public cloud object stores like Amazon S3 and Microsoft Azure have become “go-to” solutions for storing large volumes of structured and unstructured data at cloud-scale. Object storage cost-effectively provides massive scale, is designed for unstructured data and new data types generated by sensors and IoT (Internet of Things), and provides easy accessibility to stored data.
The ActiveScale P100 provides the benefits of object storage in a modular turnkey system that is easy to scale, deploy, manage and grow. The vertically integrated software, networking, storage and server rack level system is up-and-running without the challenges and hidden costs associated with assembling a similar system. The integrated components have been tuned for optimal performance, and the built-in predictive analytics of ActiveScale SM enables proactive management. With an entry-level price point and easy expandability, the ActiveScale P100 enables businesses of all sizes and in any industry to take advantage of cloud-scale data storage, easy data access, and big data analytics for a range of workloads.