1space Capabilities
A Single, Multi-Cloud NamespaceSwiftStack 1space creates a single namespace across multiple clouds—public and private. A single namespace enables flexibility for users and applications, as data location can be managed without the need to update application or user configuration. As data moves throughout the namespace, it is transparent to anyone or anything that uses it. SwiftStack 1space automates data-intensive workflows to optimize data placement and access. Policies can be customized to enable the desired workflows for your applications and users. SwiftStack 1space stores data using cloud-native formats and not in a proprietary archive. This means that users and applications can access and operate on the data directly in a public cloud. |
Mirroring Data Between Private and Public CloudsMany workflows need an additional copy of data to be stored in a different physical location for even higher levels of data availability or durability. The mirroring functionality of SwiftStack 1space is designed to address this functionality, as data can automatically be synchronized with another storage target. When a 1space profile is configured to sync/mirror data, it provides immediate, asynchronous replication to another storage location to ensure that a second copy of the data is stored offsite in a public cloud. The 1space profile can be configured to not propagate on-premises deletes to data synchronized to a public cloud. This allows data to be protected against accidental or unintended deletes. |
Lifecycling Data To and From a Public CloudSwiftStack 1space can be used in an archiving strategy by automatically moving data stored on-premises to a public cloud. Data is moved after an operator-defined period of time, configurable in days. This can be used to free physical storage capacity on-premises or to use a public cloud as another storage tier. When a 1space profile is configured to lifecycle, data will be moved to the public cloud after operator-defined number of days. There is an option for the data to be copied to the public cloud immediately, then it would be removed on-premises after the defined time period. The 1space profile can be configured to create a copy of the data on-premises if it is accessed when it only exists in the public cloud. After the data is restored and the lifecycle time is reached again, it will only exist in the public cloud. |
Placing Data Based on Object MetadataFor many workflows, not all data in the namespace needs to be moved or copied to the public cloud. For example, compute-intensive workflows may need a subset of the data in the public cloud for processing. Another use is to lifecycle completed projects to a lower-cost storage tier, or identify important data that should be mirrored to the public cloud. 1space enables sophisticated configuration of policies to be triggered by applications or users as they tag data with metadata. Watch the demo video to learn more. |
Enable Live Application Migration Between CloudsThe SwiftStack 1space migration capability can be used to execute a live migration of data stored in a public cloud to a private cloud. Even before the data is moved from the public cloud to a SwiftStack cluster, it is immediately visible to applications and users in the single namespace. This means that applications can access and utilize data as if it was already migrated. The migration activities are distributed across multiple nodes within the SwiftStack cluster to maximize the use of available network resources. While the migration is in progress, it can be monitored in the SwiftStack Controller. |
1space Architecture
1space integrated with SwiftStack1space is integrated into SwiftStack and provides a centralized data placement and policy engine to create a single addressable data storage namespace across private and public clouds. No additional software needs to be installed to use 1space, and every SwiftStack license includes all 1space capabilities. Additional functionality is added in optional “middleware” or in the SwiftStack Controller, keeping the core pure. |
Shared Nothing Architecture1space is architected with a shared-nothing approach to deployment and policy execution. This means that actors participating in the single namespace, or applying policies, can be independent and do not need centralized coordination. This is done for scale, performance and reliability. Linear scaling characteristics are realized when adding worker nodes to the 1space cluster. There is no distributed database, central coordination server, master election, metadata servers, etc. 1space enables multi-part and parallel data transfers to ensure maximum performance from each node in the cluster. Transfers are fully distributed across multiple nodes to maximize available network bandwidth. 1space clustering enables high availability and high reliability policy execution when there are 1space node failures or network partitions. |
1space ProjectAt SwiftStack, our belief is that when data is managed using open-source technologies and accessed using open protocols, it maximizes flexibility and leverages the efforts of a broader community. The technology that powers SwiftStack 1space is contributed to the open community as its own project with the current goal of helping OpenStack Swift environments move data to and from S3-compatible clouds. Learn more about the 1space project. There are four unique roles in a SwiftStack cluster (Proxy, Account, Container, and Object). Each node in the cluster is assigned one or more specific roles. For small clusters, it’s common that each node participates in each role. In larger scale clusters, some roles are given their own nodes. No matter how the cluster is initially configured, it can be seamlessly modified and scaled at any time. Both capacity and performance of a SwiftStack cluster can scale independently. To scale overall performance by allowing more concurrent connections, add nodes with the Proxy role. To scale capacity, add nodes with the Object role. For most applications—whether servers, network, or storage—adding hardware infrastructure is a tedious process. With SwiftStack, the expansion of the cluster can be automated by setting up hardware profiles for each type of server in the cluster. For example, some servers may be high-capacity for a specific use, while others may have a higher ratio of CPU for performance applications. These can be recognized automatically and added to the cluster with no administration overhead. Learn more. |
The 1space Cloud Connector is deployed as a container in the public cloud and can be provisioned as a load-balanced cluster for high availability. While data is always directly accessible in the public cloud via native APIs, the 1space Connector is an optional component for locally accessing data in the single namespace. The 1space Cloud Connector enables applications to burst to the public cloud and retain access to all data . With the 1space Cloud Connector, applications can be live-migrated to a public cloud while retaining access to on-premises data . |
1space ManagementThe SwiftStack Controller provides a simple GUI management framework for all policy creation and monitoring for 1space. Once you associate your public cloud accounts with SwiftStack, all data placement and movement policies are controlled and monitored in the SwiftStack interface. 1space profiles define data policy execution. This enables multiple storage accounts to share the same 1space execution settings. 1space mappings assign profiles to link specific on-premises storage locations with public cloud storage locations. 1space migration configures migration policies from a remote storage locations to a SwiftStack cluster. The 1space Cloud Connector actively manages the configuration of remote cloud connectors running in the public cloud as new 1space mappings are created. 1space can optionally be configured with a metadata index to support metadata search queries via an API or the SwiftStack Client. |
Metadata
1space MetadataAs we eliminate data silos, metadata becomes the new organizing force for data. Any amount of metadata can be associated with objects, which allows applications to optimize data for end-to-end workflows. Metadata can be leveraged for data discovery and data placement across on-premises and public clouds. SwiftStack can automatically index metadata associated with objects for user/application metadata search. Metadata can also be leveraged for dynamically assigning 1space policies for custom workflows. SwiftStack supports the ability to automate the tagging of metadata as data is uploaded. Additionally, metadata can be validated as data is uploaded. Custom metadata schemas can be defined to ensure consistent metadata when automating workflows. In addition to static policies, 1space can use object metadata to manage multi-cloud data placement. This is incredibly important, because we have found that the best products and services leverage the data they collect to its fullest. In some workflows, we want to collect a lot of data, but we don’t want to send *everything* to the cloud. With 1space—users or applications can create a set of tags that will trigger data movement. |
Once metadata is indexed, data can be searched by applications to build custom sets of data for data-intensive workflows. SwiftStack integrates with Elasticsearch, the leading open source indexing and search platform. As metadata changes, SwiftStack will ensure that the metadata in the search index stays up-to-date. |
SwiftStack Client includes support for metadata search. Users can collaborate with simple or complex queries that can be managed and shared through a GUI interface. SwiftStack Client can be used to set metadata that can trigger a 1space data management policy. This enables users to manage data placement with custom workflows. |
