Solano Labs, Inc.

Steven R Willis

Christopher A Thorpe

Jay Moorthi

William Josephson

Thomas E Westberg

A distributed data storage system is provided for offering shared data to one or more clients. In various embodiments, client systems operate on shared data while having a unique writeable copy of the shared data. According to one embodiment, the data storage system can be optimized for various use cases (e.g., read-mostly where writes to shared data are rare or infrequent (although writes to private data may be frequent. Some implementations of the storage system are configured to provide fault tolerance and scalability for the shared storage. For example, read-only data can be stored in (relatively) high latency, low cost, reliable storage (e.g. cloud), with multiple layers of cache supporting faster retrieval. In addition, some implementations of the data storage system offer a low-latency approach to data caching. Other embodiments improve efficiency with access modeling and conditional execution cache hints that can be distributed across the data storage system.

Systems and methods for efficient distribution of stored data objects

Systems and methods are provided for managing and/or executing cloud compute instances that may be pre-empted by their providers, known in at least one provider as 'spot instances'. Various aspects manage early termination (i.e., before customer deallocation or compute task completion) of spot instances, enabling improved utilization, and increasing reliability of executing tasks on spot instances. In some embodiments, systems and/or methods are provided for prediction of when spot instance kills are likely to occur. These systems and/or methods can gracefully handle spot kills issued by Providers (e.g., trigger preservation operations), trigger transitions to spot instances at another Provider, trigger increased bidding to preserve current spot instances, and/or trigger a premium payment to allow for execution of preservation operations, among other options. Other embodiments enable customers to directly manage instance kill operations (e.g., a customer or a customer's program selects instances to terminate via a UI or API, etc.).

Growing community of inventors

Cambridge, MA, United States of America

Steven R Willis