A Read-Write-Validate approach to optimistic concurrency control for energy efficiency of resource-constrained systems

Abstract

Modern smartphones feature multiple applications which access shared data on the solid state storage within the device. As applications become more complex, contention over this memory resource is becoming an issue. This leads to increased battery drain as the applications are forced to touch the solid state device repeatedly after failing to retrieve or store data due to contention from other applications. We describe an optimistic concurrency control algorithm, combining a novel Read-WriteValidate phase sequence with virtual execution. The protocol is suitable for governing transactions operating on databases residing on resource-constrained devices. Increasing energy efficiency and reducing latency are primary goals for our algorithm. We show that this is achieved by reducing persistent store access, and satisfy realtime requirements via transaction scheduling that affords greater determinism. AbstractModern smartphones feature multiple applications which access shared data on the solid state storage within the device. As applications become more complex, contention over this memory resource is becoming an issue. This leads to increased battery drain as the applications are forced to touch the solid state device repeatedly after failing to retrieve or store data due to contention from other applications. We describe an optimistic concurrency control algorithm, combining a novel Read-Write-Validate phase sequence with virtual execution. The protocol is suitable for governing transactions operating on databases residing on resource-constrained devices. Increasing energy efficiency and reducing latency are primary goals for our algorithm. We show that this is achieved by reducing persistent store access, and satisfy real-time requirements via transaction scheduling that affords greater determinism. About the authorsKamal Solaiman is a Phd student in the School of Computing Science at Newcastle University. Abstract -Modern smartphones feature multiple applications which access shared data on the solid state storage within the device. As applications become more complex, contention over this memory resource is becoming an issue. This leads to increased battery drain as the applications are forced to touch the solid state device repeatedly after failing to retrieve or store data due to contention from other applications. We describe an optimistic concurrency control algorithm, combining a novel Read-Write-Validate phase sequence with virtual execution. The protocol is suitable for governing transactions operating on databases residing on resource-constrained devices. Increasing energy efficiency and reducing latency are primary goals for our algorithm. We show that this is achieved by reducing persistent store access, and satisfy real-time requirements via transaction scheduling that affords greater determinism. Matthew Brook is a Phd student in the

Optimistic Concurrency Control for Energy Efficiency in the Wireless Environment

“…In [8] we identified the possibility that, in combination with virtual execution, a performance improvement could be made by allowing the write phase to be accomplished before the validation phase . In [8] [14] we explored this idea in the context of multiple applications running concurrently on a resource-constrained device.…”

“…In [8] we applied this thinking to concurrent transactions on the shared resources of a smart device. The use of a virtual execution enabled OCC coupled with the reordering of the validation and write phases allowed for an overall improvement in performance.…”

“…We achieve this by changing the ordering of the validate step at the server so that it takes place after the write step (conventionally it occurs before writing). In [8] we showed that this approach is applicable as optimistic concurrency control on resource-constrained devices such as smart-phones. We now extend this work to the broadcast disk model for mobile network applications that require significantly more read transactions than write transactions.…”