Abstract: Systems and methods are provided for management of data access latency in a distributed storage system by dynamically replicating data objects experiencing a high request volume. When a storage node is unable to meet a data access latency target (e.g., due to one or more data objects hosted by the storage node being the subject of a large quantity of requests), the storage node may initiate a process by which a second storage node is employed to reduce the request processing load. The second storage node may cache a copy of the requested data object, and may service at least a portion of requests for the data object using the cached copy.

Abstract: Systems and methods are provided for implementing a multi-service file system for a hosted computing instance via a locally-addressable secure compute layer. Software within the instance can submit file operations to the secure compute layer, which the secure compute layer can translate into calls to one or more network-accessible storage services. To provide a multi-service file system, the secure compute layer can obtain mapping data mapping file system objects within the virtualized file system to different network-accessible storage services. On receiving a file operation, the secure compute layer can determine one or more network-accessible storage services corresponding to the file operation, and submit appropriate calls to the one or more network-accessible storage services.

Abstract: Systems and methods are provided for exposing disparate storage systems and technologies to compute instances through a unified file system interface that is agnostic to the underlying storage systems and technologies. A file system namespace service may provide the ability to compose a file system from different storage technologies and systems (e.g., block storage, object storage, etc.). The file system namespace service may also provide the ability to layer different storage services (e.g., by treating local block storage volume as a write back cache, by treating a file system as a read-through cache in front of an object storage service, etc.), and/or the ability to bind virtual files to other compute instances and/or storage systems.

Abstract: Sets of asynchronous replication operations may be tracked to ensure consistency. A tracking service may receive notifications of pending asynchronous replication tasks, and responsive to receiving a manifest indicating a request to be notified upon completion of a set of pending replication asynchronous tasks, matches individual ones of the tasks within the set to tasks indicated as pending. The tracking service may then select a routing table based on a most recent sequence number associated with the set of tasks, determine one of more tracking nodes assigned to track the set of tasks, and send the manifest to each of the tracking nodes. As individual ones of the tasks complete, notifications of completion may be sent to the tracking nodes and an aggregator node aggregates the completion notifications for the set. Once all completion notifications are received, a response to the request indicating completion may be sent.

Abstract: Dynamic management of data traffic workflows is performed. An event to perform a data traffic workflow at a remote performance location may be received. Computing resources to perform the data traffic workflow may be identified. Operations to perform the data traffic workflow may be dynamically directed by the identified computing resources to adaptively balance performance of the operations with operations for other data traffic workflows in order to meet respective performance requirements of the data traffic workflows.

Abstract: The performance of asynchronous tasks may be dynamically configured. An evaluation of pending tasks that includes updates for new tasks and completed tasks being removed is performed. A target capacity for the data traffic workflow is determined from the evaluation of pending tasks in order to satisfy a performance requirement for the data traffic workflow. Modifications to a performance configuration for the data traffic workflow are then made based on a comparison with the target capacity.

Abstract: A data transport service of a provider network provide isolated and flexible network data transmission between different computing infrastructure collections of the provider network (e.g., between different geographical regions) for different users. The data transport service may receive a request from a user for assignment of a channel. The request specifies a desired capacity of the channel to transmit data from a computing infrastructure collection to another a computing infrastructure collection. Based on the requested capacity for the channel, the data transport service may reserve, for the channel, network resource capacity to transmit the data between the infrastructures and reserve, for the channel, compute instance capacity to ingest the data and to send the data between the infrastructures using the reserved network resource capacity. The reserved network resource capacity and reserved compute instance capacity are dedicated to the channel and are unavailable for use by other channels.