ScoutAM
EXABYTE-SCALE DATA ORCHESTRATION
Versity’s all-new ScoutAM platform is the first scalable, modular, economically efficient mass storage solution to address the need for a simple but powerful tool to manage and preserve exponentially increasing collections of unstructured data.
ScoutAM benefits from the rich legacy of VSM and the products that came before it – namely SAM-QFS. Over the past eight years, there have been many engineering lessons learned, which have all been incorporated into the ScoutAM platform. Although the code and architecture are new, many years of experience have informed the design and engineering choices behind the product.
While ScoutAM was developed for exascale workloads, it was also designed for ease of use, configuration, and deployment. Our vision for the product was to create the most capable mass storage platform in the world while ensuring that it is easy to deploy and manage.
ScoutAM is the first mass storage system designed for ease of use and quick startup with a simple installation and configuration that can be completed in as little as 30 minutes
How it works
Data is ingested from primary sources to the ScoutAM cache over industry-standard file and object protocols. Metadata is recorded and separated from data where it remains permanently online. Data resides in the cache for a defined time interval or until the storage space is needed for new data. Automated policies are applied to cache data to generate the desired number of copies and write the copies to the defined destination. Policy definitions enable grouping by data type, application, size, user, or group. These definitions also set boundaries on the time that can pass before the creation of copies and boundaries on the data set size. The ability to apply different policies and coalesce random incoming data streams into efficient streaming data sets is one of the platform’s core functions.
SCOUTAM FEATURES
Graphical User Interface
ScoutAM includes a modern API-driven graphical user interface to administer the system, monitoring, alerting, and configuration.
Parallel Transfer
ScoutAM can read and write data to mass storage devices and services through multiple data channels on each node in the cluster simultaneously. Large files or objects may be segmented and written in parallel across a configurable number of data channels or a range of data channels. Many smaller files or objects may be scheduled across channels using a round-robin algorithm.
High Availability
ScoutAM remains available despite the loss of servers in a cluster. Depending on the cluster’s size and the quorum definitions, ScoutAM can tolerate the loss of one or more servers with no impact on availability or continuity of services. Failover is built into the ScoutAM platform and does not require complex external failover or HA tools.
Replication
ScoutAM supports asynchronous read-only remote replication. Metadata, cache data, and mass storage or archival data may be replicated to one or more disaster recovery or secondary locations to ensure continuity of operations.
Migration Free
ScoutAM is the first mass storage platform to support data formats from third party platforms including; DMF, HPSS, & OHSM. Format support eliminates the need for costly and time consuming data migrations.
Extended Cache
ScoutAM supports the optional ability to increase cache capacity by adding low-cost S3 object storage devices to augment the capacity of the primary cache. The space available on the extended cache will be fully utilized, and the least used data will be automatically evicted as newer data arrives. Pairing extended cache storage with an all-flash primary cache enables an optimal mix of performance and capacity at a lower TCO.
Hybrid Cloud
ScoutAM supports both on premises S3 object storage devices and Amazon, Azure, and Google public cloud storage services. High performance cloud capabilities enable Versity sites to benefit from the flexibility of hybrid tape and cloud infrastructure.
Automated Orchestration
ScoutAM supports the creation and automatic execution of site specific data orchestration policies.
Open Source Format
Open-source metadata combined with open-source data formats gives customers complete control over data collections and aligns with long-term data preservation and autonomy goals.
Object to Tape
ScoutAM is capable of ingesting S3 object data though a scale out Versity S3 gateway and storing the objects data to tape for long term protection.
Ultimate Disaster Recovery
ScoutAM supports recovery of all data elements directly from the mass storage media, without the use of any specialized software. All information required to restore a data collection resides on the physical media with the data.
API Coverage
The ScoutAM platform includes extensive API coverage to ease integration with third party tools and site specific workflows. All elements of the ScoutAM GUI are accessed by published API’s.
Modular
ScoutAM is a modular platform that can be expanded incrementally to increase capacity and performance.
ScoutAM Architecture
Introducing the ScoutAM Nodes and Services Architecture.
Each server in a ScoutAM cluster can deliver all of the platform services and scales modularly to increase total performance. This architecture ensures availability by eliminating single points-of-failure and delivers scalability by removing bottlenecks for metadata processing and parallel data movement between primary and mass storage environments.
The ScoutAM runtime application is present on all nodes in the cluster. Each node runs a ScoutAM executor, and the cluster runs a ScoutAM schedule – a highly scalable virtual role that is not dependent on any specific server in the cluster. Each ScoutAM node can deliver data through as many data channels as the server configuration allows and will deliver approximately 10 GB/s of aggregate throughput per node for mid-range servers.
— Senior Data Protection Engineer, Global Banking Customer