Journal of the ACM (JACM)
My Cache or Yours? Making Storage More Exclusive
ATEC '02 Proceedings of the General Track of the annual conference on USENIX Annual Technical Conference
Calculating stack distances efficiently
Proceedings of the 2002 workshop on Memory system performance
Empirical evaluation of multi-level buffer cache collaboration for storage systems
SIGMETRICS '05 Proceedings of the 2005 ACM SIGMETRICS international conference on Measurement and modeling of computer systems
Evaluation techniques for storage hierarchies
IBM Systems Journal
Proceedings of the 8th international conference on Mobile systems, applications, and services
TaintDroid: an information-flow tracking system for realtime privacy monitoring on smartphones
OSDI'10 Proceedings of the 9th USENIX conference on Operating systems design and implementation
IOMMU: strategies for mitigating the IOTLB bottleneck
ISCA'10 Proceedings of the 2010 international conference on Computer Architecture
Full-system analysis and characterization of interactive smartphone applications
IISWC '11 Proceedings of the 2011 IEEE International Symposium on Workload Characterization
Rigorous rental memory management for embedded systems
ACM Transactions on Embedded Computing Systems (TECS) - Special section on ESTIMedia'12, LCTES'11, rigorous embedded systems design, and multiprocessor system-on-chip for cyber-physical systems
Hi-index | 0.00 |
Most embedded systems require contiguous memory space to be reserved for each device, which may lead to memory under-utilization. Although several approaches have been proposed to address this issue, they have limitations of either inefficient memory usage or long latency for switching the reserved memory space between a device and general-purpose uses. Our scheme utilizes the reserved memory as an eviction-based file cache. It guarantees contiguous memory allocation to devices while providing idle device memory as an additional file cache called eCache for general-purpose usage. Since the eCache stores only evicted data from in-kernel page cache, memory efficiency is preserved and allocation time for devices is minimized. Cost-based region selection also alleviates additional read I/Os by carefully discarding cached data from the eCache. The prototype is implemented on the Nexus S smartphone and is evaluated with popular Android applications. The evaluation results show that 50%-85% of flash read I/Os are reduced and application launch performance is improved by 8%-16% while the reallocation time is limited to a few milliseconds.