Self-adjusting Advertisement of Cache Indicators with Bandwidth Constraints

Cohen, Itamar; Einziger, Gil; Scalosub, Gabriel

doi:10.1109/infocom42981.2021.9488680

Cited by 3 publications

(2 citation statements)

References 43 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Since constantly advertising a fresh indicator might be prohibitively costly, in practice caches commonly advertise fresh indicators only periodically [2], [16], [19]- [21], where one usually refers to the period between the advertisements of fresh indicators as the update interval. Several works address the interplay between the update interval and performance by means of simulations [2], [16].…”

Section: B Related Workmentioning

confidence: 99%

“…Several works address the interplay between the update interval and performance by means of simulations [2], [16]. The work [21] presents an algorithm that dynamically scales the update interval and the indicator size, to comply with bandwidth constraints. The works [22], [23] reduce the transmission overheads by accurately advertising important information, while allowing less important information to be stale, or less accurate.…”

Section: B Related Workmentioning

confidence: 99%

See 1 more Smart Citation

False Negative Awareness in Indicator-based Caching Systems

Cohen¹,

Einziger²,

Scalosub³

2022

Preprint

Self Cite

View full text Add to dashboard Cite

Distributed caching systems such as content distribution networks often advertise their content via lightweight approximate indicators (e.g., Bloom filters) to efficiently inform clients where each datum is likely cached. While false-positive indications are necessary and well understood, most existing works assume no false-negative indications. Our work illustrates practical scenarios where false-negatives are unavoidable and ignoring them significantly impacts system performance. Specifically, we focus on false-negatives induced by indicator staleness, which arises whenever the system advertises the indicator only periodically, rather than immediately reporting every change in the cache. Such scenarios naturally occur, e.g., in bandwidthconstraint environments or when latency impedes each client's ability to obtain an updated indicator. Our work introduces novel false-negative aware access policies that continuously estimate the false-negative ratio and sometimes access caches despite negative indications. We present optimal policies for homogeneous settings and provide approximation guarantees for our algorithms in heterogeneous environments. We further perform an extensive simulation study with multiple real system traces. We show that our false-negative aware algorithms incur a significantly lower access cost than existing approaches or match the cost of these approaches while requiring an order of magnitude fewer resources (e.g., caching capacity or bandwidth).

show abstract

Section: B Related Workmentioning

confidence: 99%