In a chip-multiprocessor (CMP) system, the DRAM system isshared among cores. shared requests from athread can not only delay other threads by causingbank/bus/row-buffer conflicts but they also destroy threads’DRAM-bank-level parallelism. Requests whose latencies would otherwisehave been overlapped could effectively become serialized. As aresult both fairness and throughput degrade, some threadscan starve for long time periods.This paper proposes fundamentally new approach to designinga...

Energy has become a first-class design constraint in computer systems. Memory is significant contributor to total system power. This paper introduces Flikker, an application-level technique reduce refresh power DRAM memories. Flikker enables developers specify critical and non-critical data programs the runtime allocates this separate parts of memory. The portion memory containing refreshed at regular refresh-rate, while substantially lower rates. partitioning saves energy cost modest...

Networking over UHF white spaces is fundamentally different from conventional Wi-Fi along three axes: spatial variation, temporal and fragmentation of the spectrum. Each these differences gives rise to new challenges for implementing a wireless network in this band. We present design implementation Net7, first like system constructed on top spaces. Net7 incorporates adaptive spectrum assignment algorithm handle variation fragmentation, proposes low overhead protocol variation. builds simple...

Buffers in on-chip networks consume significant energy, occupy chip area, and increase design complexity. In this paper, we make a case for new approach to designing interconnection that eliminates the need buffers routing or flow control. We describe algorithms without using router input/output ports. analyze advantages disadvantages of bufferless discuss how latency can be reduced by taking advantage fact do not exist. Our evaluations show significantly reduces energy consumption...

We define and study the scheduling complexity in wireless networks, which expresses theoretically achievable efficiency of MAC layer protocols. Given a set communication requests arbitrary describes amount time required to successfully schedule all requests. The most basic important network structure networks being connectivity, we i.e., minimal until connected can be scheduled. In this paper, prove that connectivity grows only polylogarithmically number nodes. Specifically, present novel...

In a chip-multiprocessor (CMP) system, the DRAM system is shared among cores. requests from thread can not only delay other threads by causing bank/bus/row-buffer conflicts but they also destroy threadspsilaDRAM-bank-level parallelism. Requests whose latencies would otherwise have been overlapped could effectively become serialized. As are sult both fairness and throughput degrade, some scan starve for long time periods. This paper proposes fundamentally new approach to designing controller...

Main memory is a major shared resource among cores in multicore system. If the interference between different applications' requests not controlled effectively, system performance can degrade significantly. Previous work aimed to mitigate problem of applications by changing scheduling policy controller, i.e., prioritizing from way that benefits performance.

The 2010 FCC ruling on white spaces proposes relying a database of incumbents as the primary means determining space availability at any device (WSD). While provides broad guidelines for database, specifics its design, features, implementation, and use are yet to be determined. Furthermore, architecting network where all WSDs rely raises several systems networking challenges that have remained unexplored. Also, treats only storehouse incumbents. We believe mandated has an additional...

The question of what can be computed, and how efficiently, is at the core computer science. Not surprisingly, in distributed systems networking research, an equally fundamental computed a fashion. More precisely, if nodes network must base their decision on information local neighborhood only, well they compute or approximate global (optimization) problem? In this paper we give first polylogarithmic lower bound such computation for problems including minimum vertex cover, (connected)...

As an innovative mobility strategy, public bike-sharing has grown dramatically worldwide. Though providing convenient, low-cost and environmental-friendly transportation, the unique features of systems give rise to problems both users operators. The primary issue among these is uneven distribution bicycles caused by ever-changing usage (available) supply. This bicycle imbalance necessitates efficient bike re-balancing strategies, which depends highly on modeling prediction. In this paper,...

We give time lower bounds for the distributed approximation of minimum vertex cover (MVC) and related problems such as dominating set (MDS). In k communication rounds, MVC MDS can only be approximated by factors Ω(nc/k2/k) Ω(Δ>1/k/k) some constant c, where n Δ denote number nodes largest degree in graph. The rounds required order to achieve a or even polylogarithmic ratio is at least Ω(√log n/log log n) Ω(logΔ/ Δ). By simple reduction, latter also hold construction maximal matchings independent sets.

A number of studies have shown the abundance unused spectrum in TV bands. This is stark contrast to overcrowding wireless devices ISM recent trend alleviate this disparity design Cognitive Radios, which constantly sense and opportunistically utilize frequencies In paper, we introduce concept a time-spectrum block model reservation, use it present theoretical formalization allocation problem cognitive radio networks. We centralized distributed protocol for show that these protocols are close...

We study a fundamental yet under-explored facet in wireless communication -- the width of spectrum over which transmitters spread their signals, or channel width. Through detailed measurements controlled and live environments, using only commodity 802.11 hardware, we first quantify impact on throughput, range, power consumption. Taken together, our findings make strong case for systems that adapt Such adaptation brings unique benefits. For instance, when throughput required is low, moving to...

To date, topology control in wireless ad hoc and sensor networks--the study of how to compute from the given communication network a subgraph with certain beneficial properties .has been considered as static problem only; time required actually schedule links computed without message collision was generally ignored. In this paper we analyze context physical Signal-to-Interference-plus-Noise-Ratio (SINR) model, focusing on question fast resulting can be realized over time.For purpose, define...

Network-on-Chips (NoCs) are likely to become a critical shared resource in future many-core processors. The challenge is develop policies and mechanisms that enable multiple applications efficiently fairly share the network, improve system performance. Existing local packet scheduling routers fail fully achieve this goal, because they treat every equally, regardless of which application issued packet.

Achieving a global goal based on local information is challenging, especially in complex and large-scale networks such as the Internet or even human brain. In this paper, we provide an almost tight classification of possible trade-off between amount quality solution for general covering packing problems. Specifically, give distributed algorithm using only small messages which obtains (ρΔ)1/k-approximation problems time O(k2), where ρ depends LP's coefficients. If message size unbounded,...

FCC has proposed to allow unlicensed operations in the TV broadcast bands. This portion of spectrum several desirable properties for more robust data communication as compared ISM However, there are a number challenges efficiently using For example, available is fragmented, and its availability may vary over time. In this paper, we present cognitive radio system, called KNOWS, address these challenges. We design our prototype, which includes new hardware platform spectrum-aware Medium Access...

A newly deployed multi-hop radio network is unstructured and lacks a reliable efficient communication scheme. In this paper, we take step towards analyzing the problems existing during initialization phase of ad hoc sensor networks. Particularly, model as quasi unit disk graph allow nodes to wake up asynchronously at any time. Further, do not feature collision detection mechanism, they have only limited knowledge about topology. We show that even for restricted model, good clustering can be...

We study a fundamental yet under-explored facet in wireless communication -- the width of spectrum over which transmitters spread their signals, or channel width. Through detailed measurements controlled and live environments, using only commodity 802.11 hardware, we first quantify impact on throughput, range, power consumption. Taken together, our findings make strong case for systems that adapt Such adaptation brings unique benefits. For instance, when throughput required is low, moving to...

DRAM is facing severe scalability challenges in sub-45nm tech- nology nodes due to precise charge placement and sensing hur- dles deep-submicron geometries. Resistive memories, such as phase-change memory (PCM), already scale well beyond are a promising replacement. Unfortunately, PCM write-limited, current approaches managing writes must de- commission pages of when the first bit fails. This paper presents dynamically replicated (DRM), hardware operating system interface designed for that...

The key application scenario of wireless sensor networks is data gathering nodes transmit data, possibly in a multi-hop fashion, to an information sink. performance thus characterized by the rate at which can be aggregated In this paper, we derive first scaling laws describing achievable worst-case i.e.arbitrarily deployed,sensor networks. We show that physical model communication and for large number practically important functions, sustainable Ω(1 / log2 n) achieved every network even when...