Abstract: The present disclosure provides for maintaining a steady load on a global database. In a global database with regionalized replication, some requests are handled by a global clone in a region, as opposed to the global database. If the global clone in the region is not up to date, the request may “fallback,” where it is redirected to the global database. By serving some requests at the global clone, instead of the global database, the traffic to the global data may be significantly reduced. However, the global database has different mechanisms which depend on a steady load. Accordingly, the present disclosure provides a mechanism for maintaining a constant load on the global database, despite regionalized replication. Maintaining a constant load ensures that the global database will be able to handle fallbacks when request to clones fail.

Abstract: The present disclosure provides for consistent reads in a global database that is replicated to one or more global clones. A state machine is implemented, and executed independently, for each project in each region. Based on the state of a project when a read request is received, the read request will be served by the global database, served by the global clone, or ignored as an invalid request.

Abstract: The present disclosure provides for consistent reads in a global database that is replicated to one or more global clones. A state machine is implemented, and executed independently, for each project in each region. Based on the state of a project when a read request is received, the read request will be served by the global database, served by the global clone, or ignored as an invalid request.

Abstract: Techniques for selecting a network communications protocol based on network topology and/or network performance are described. A first application executing on a first computer system obtains a distance characteristic of a network coupling the first computer system to a second computer system. The first application selects a network communications protocol from a plurality of network communications protocols based on the distance characteristic of the network. The first application and the second application connect with the selected network communications protocol. In some embodiments, the distance characteristic of the network may be based on topological or performance characteristics of the network.

Abstract: The present disclosure provides for maintaining a steady load on a global database. In a global database with regionalized replication, some requests are handled by a global clone in a region, as opposed to the global database. If the global clone in the region is not up to date, the request may “fallback,” where it is redirected to the global database. By serving some requests at the global clone, instead of the global database, the traffic to the global data may be significantly reduced. However, the global database has different mechanisms which depend on a steady load. Accordingly, the present disclosure provides a mechanism for maintaining a constant load on the global database, despite regionalized replication. Maintaining a constant load ensures that the global database will be able to handle fallbacks when request to clones fail.

Abstract: The present disclosure provides for consistent reads in a global database that is replicated to one or more global clones. A state machine is implemented, and executed independently, for each project in each region. Based on the state of a project when a read request is received, the read request will be served by the global database, served by the global clone, or ignored as an invalid request.

Abstract: Generally described, one or more aspects of the present application correspond to techniques for new block store volumes of data from object storage snapshots of an origin volume by creating an intermediary “acceleration pool” of source volumes reserved for supporting the new volume creation process. The source volumes may be prevented from supporting user I/O or computing instances in order to reserve their bandwidth for the transfer of data to the new volumes. In addition, the source volumes can be stored on block store servers to increase the speed with which their data can be transferred to the new volumes. The acceleration pool can be partitioned among users accessing the pool in order to maintain availability even when some users heavily utilize the acceleration pool, with particular source volumes intelligently selected from among a user's partition in order to optimize the speed at which the new volumes are created.

Abstract: A first request to create a first storage volume to store a first portion of a first logical volume is received by a first block storage server instance. A second request to create a second storage volume to store a second portion of the first logical volume is received by a second block storage server instance. A third request is sent to a third block storage server instance to create a third storage volume to store the second portion of the first logical volume. The second portion of the first logical volume is stored by the third block storage server instance to the third storage volume. A data store containing an identification of each block storage server instance hosting a portion of the first logical volume is updated to remove an identification of the second block storage server instance and add an identification of the third block storage server instance.

Abstract: Generally described, one or more aspects of the present application correspond to techniques for new block store volumes of data from object storage snapshots of an origin volume by creating an intermediary “acceleration pool” of source volumes reserved for supporting the new volume creation process. The source volumes may be prevented from supporting user I/O or computing instances in order to reserve their bandwidth for the transfer of data to the new volumes. In addition, the source volumes can be stored on block store servers to increase the speed with which their data can be transferred to the new volumes. The acceleration pool can be partitioned among users accessing the pool in order to maintain availability even when some users heavily utilize the acceleration pool, with particular source volumes intelligently selected from among a user's partition in order to optimize the speed at which the new volumes are created.

Abstract: A first request to create a first storage volume to store a first portion of a first logical volume is received by a first block storage server instance. A second request to create a second storage volume to store a second portion of the first logical volume is received by a second block storage server instance. A third request is sent to a third block storage server instance to create a third storage volume to store the second portion of the first logical volume. The second portion of the first logical volume is stored by the third block storage server instance to the third storage volume. A data store containing an identification of each block storage server instance hosting a portion of the first logical volume is updated to remove an identification of the second block storage server instance and add an identification of the third block storage server instance.