Dmitrii Ustiugov

  • Citations Per Year
Learn More
Modern in-memory services rely on large distributed object stores to achieve the high scalability essential to ser-vice thousands of requests concurrently. The independent and unpredictable nature of incoming requests results in random accesses to the object store, triggering frequent remote memory accesses. State-of-the-art distributed memory frameworks(More)
The increasing demand for extracting value out of ever-growing data poses an ongoing challenge to system designers, a task only made trickier by the end of Dennard scaling. As the performance density of traditional CPU-centric architectures stagnates, advancing compute capabilities necessitates novel architectural approaches. Near-memory processing (NMP)(More)
In this report, we describe a number of Hardware Transactional Memory (HTM) designs and their basic mechanisms used for better programmability and higher performance than conventional synchronization techniques based on locking. We compare the systems considering their programming model, hardware design challenges, transactional dataset constraints and(More)
  • 1