Optimal Partitioning of Cache Memory
IEEE Transactions on Computers
Packet classification on multiple fields
Proceedings of the conference on Applications, technologies, architectures, and protocols for computer communication
Characterizing locality, evolution, and life span of accesses in enterprise media server workloads
NOSSDAV '02 Proceedings of the 12th international workshop on Network and operating systems support for digital audio and video
Dynamic Partitioning of Shared Cache Memory
The Journal of Supercomputing
A page fault equation for modeling the effect of memory size
Performance Evaluation
A new approach to dynamic self-tuning of database buffers
ACM Transactions on Storage (TOS)
Proceedings of the 5th international conference on Emerging networking experiments and technologies
METE: meeting end-to-end QoS in multicores through system-wide resource management
ACM SIGMETRICS Performance Evaluation Review - Performance evaluation review
Evaluating per-application storage management in content-centric networks
Computer Communications
Hi-index | 0.00 |
Named Data Networking (NDN) is becoming a prominent candidate for the next Internet architecture. One of the main features of the proposal lies in using in-network caching, which makes the performance of the storage in a router very crucial. As this storage capacity is very limited, to maintain its high functionality it is better to separate the contents of some applications (i.e., video, web, etc.) which can be reused by other clients and protect them from being replaced by non-cacheable content of other applications like emails, telephony app. etc., i.e. partitioning the content store. Partitioning might also provide means for ISPs to offer different level of quality for different applications based on service level agreements with the content producers. This paper considers dividing NDN traffi into classes, and examines how an NDN router's content store can be partitioned among these classes. It illustrates how this partitioning can be done dynamically by using the Buffer Miss Equation from previous work. This illustration considers two cases with different objectives: minimizing average (over all traffi classes) miss ratio and maintaining fairness (in miss ratios) among traffi classes.