Young-ri Choi

ExScal: elements of an extreme scale wireless sensor network

Project ExScal (for extreme scale) fielded a 1000+ node wireless sensor network and a 200+ node peer-to-peer ad hoc network of 802.11 devices in a 13km by 300m remote area in Florida, USA during December 2004. In comparison with previous deployments, the ExScal application is relatively complex and its networks are the largest ones of either type fielded to date. In this paper, we overview the key requirements of ExScal, the corresponding design of the hardware/software platform and application, and some results of our experiments.

Reliable bursty convergecast in wireless sensor networks

We address the challenges of bursty convergecast in multi-hop wireless sensor networks, where a large burst of packets from different locations needs to be transported reliably and in real-time to a base station. Via experiments on a 49 MICA2 mote sensor network using a realistic traffic trace, we determine the primary issues in bursty convergecast, and accordingly design a protocol, RBC (for Reliable Bursty Convergecast), to address these issues: To improve channel utilization and to reduce ack-loss, we design a window-less block acknowledgment scheme that guarantees continuous packet forwarding and replicates the acknowledgment for a packet; to alleviate retransmission-incurred channel contention, we introduce differentiated contention control. Moreover, we design mechanisms to handle varying ack-delay and to reduce delay in timer-based retransmissions. We evaluate RBC, again via experiments, and show that compared to a commonly used implicit-ack scheme, RBC doubles packet delivery ratio and reduces end-to-end delay by an order of magnitude, as a result of which RBC achieves a close-to-optimal goodput.

The mote connectivity protocol

An attractive architecture for sensor networks is to have the sensing devices mounted on small computers, called motes. Motes are battery-powered, and can communicate in a wireless fashion by broadcasting messages over radio frequency. In mote networks, the connectivity of a mote u can be defined by those motes that can receive messages from u with high probability and those motes from which u can receive messages with high probability. In this paper, we describe a protocol that can be triggered by any mote in a mote network in order that each mote in the network computes its connectivity. The protocol is simple and has several energy saving features. We implemented this protocol over TinyOS and discuss the results of some execution runs of this implementation.

Stabilization of Grid Routing in Sensor Networks

We present a protocol for routing data messages from any sensor to the base station in a sensor network. The protocol maintains an incoming spanning tree whose root is the base station. The spanning tree is constructed as follows. First, each sensor in the network is assigned a unique identifier as if the sensors form a logical two-dimensional grid. Second, each sensor, other than the base station, uses its own identifier to compute the identifiers of its “potential parents” in the spanning tree. Third, the base station starts to periodically send “connected messages”. When a sensor receives a connected message from anyone of its potential parents, the sensor makes this potential parent its parent in the tree and starts to periodically send connected messages. This routing protocol is stabilizing such that starting from any state, the protocol converges to a state where all the sensors and only the sensors that can be connected to the routing tree are connected to the tree. The convergence time of the protocol is proportional to the diameter of the sensor network. The routing protocol also has several other advantages over earlier protocols: overhead of the protocol is small, the protocol avoids unreliable long links to build a reliable routing tree, the protocol balances the load over the whole network, and it has nice fault-tolerance property. We have evaluated this protocol over a sensor network that consisted of about 50 MICA2 motes using a realistic traffic trace, and observed that the protocol delivers 72–99% of the messages to the base station.

The effect of multi-core on HPC applications in virtualized systems

In this paper, we evaluate the overheads of virtualization in commercial multicore architectures with shared memory and MPI-based applications. We find that the non-uniformity of memory latencies affects the performance of virtualized systems significantly. Due to the lack of support for non-uniform memory access (NUMA) in the Xen hypervisor, shared memory applications suffer from a significant performance degradation by virtualization. MPI-based applications show more resilience on sub-optimal NUMA memory allocation and virtual machine (VM) scheduling. However, using multiple VMs on a physical system for the same instance of MPI applications may adversely affect the overall performance, by increasing I/O operations through the domain 0 VM. As the number of cores increases on a chip, the cache hierarchy and external memory will become more asymmetric. As such non-uniformity in memory systems increases, NUMA and cache awareness in VM scheduling will be critical for shared memory applications.

Resource Allocation Policies for Loosely Coupled Applications in Heterogeneous Computing Systems

High-Throughput Computing (HTC) and Many-Task Computing (MTC) paradigms employ loosely coupled applications which consist of a large number, from tens of thousands to even billions, of independent tasks. To support such large-scale applications, a heterogeneous computing system composed of multiple computing platforms with different types such as supercomputers, grids, and clouds can be used. On allocating heterogeneous resources of the system to multiple users, there are three important aspects to consider: fairness among users, efficiency for maximizing the system throughput, and user satisfaction for reducing the average user response time. In this paper, we present three resource allocation policies for multi-user and multi-application workloads in a heterogeneous computing system. These three policies are a fairness policy, a greedy efficiency policy, and a fair efficiency policy. We evaluate and compare the performance of the three resource allocation policies over various settings of a heterogeneous computing system and loosely coupled applications, using simulation based on the trace from real experiments. Our simulation results show that the fair efficiency policy can provide competitive efficiency, with a balanced level of fairness and user satisfaction, compared to the other two resource allocation policies.

Hop chains: secure routing and the establishment of distinct identities

We present a secure routing protocol that is immune to Sybil attacks, and that can tolerate initial collusion of Byzantine routers, or runtime collusion of non-adjacent Byzantine routers in the absence of collusion between adjacent routers. For these settings, the calculated distance from a destination to a node is not smaller than the actual shortest distance from the destination to the node. The protocol can also tolerate initial collusion of Byzantine routers and runtime collusion of adjacent Byzantine routers but in the absence of runtime collusion between non-adjacent routers. For this setting, there is a bound on how short the calculated distance is compared to the actual shortest distance. The protocol makes very weak timing assumptions and requires synchronization only between neighbors or second neighbors. We propose to use this protocol for secure localization of routers using hop-count distances, which can be then used as a proof of identity of nodes.

Project exscal

Project ExScal (for Extreme Scale) fielded a 1000+ node wireless sensor network and a 200+ node ad hoc network of 802.11 devices in a 1.3km by 300m remote area in Florida during December 2004. In several respects, these networks are likely the largest deployed networks of either type to date. We overview here the key requirements of the project, describe briefly how they were met and experimentally tested, and provide a pointer to our experimental results.

Sentries and sleepers in sensor networks

A sensor is a battery-operated small computer with an antenna and a sensing board that can sense magnetism, sound, heat, etc. Sensors in a network can use their antennas to communicate in a wireless fashion by broadcasting messages over radio frequency to neighboring sensors in the same network. In order to lengthen the relatively short lifetime of sensor batteries, each sensor in a network can be replaced by a group of n sensors, for some n ≥ 2. The group of n sensors act as one sensor, whose lifetime is about n times that of a regular sensor as follows. For a time period, only one sensor in the group, called sentry, stays awake and performs all the tasks assigned to the group, while the remaining sensors, called sleepers, go to sleep to save their batteries. At the beginning of the next time period, the sleepers wake up, then all the sensors in the group elect a new sentry for the next time period, and the cycle repeats. In this paper, we describe a practical protocol that can be used by a group of sensors to elect a new sentry at the beginning of each time period. Our protocol, unlike earlier protocols, is based on the assumption that the sensors in a group are perfectly identical (e.g. they do not have unique identifiers; rather each of them has the same group identifier). This feature makes our protocol resilient against any attack by an adversary sensor in the group that may lie about its own identity to be elected a sentry over and over, and keep the legitimate sensors in the group asleep for a long time.

Platform and Co-Runner Affinities for Many-Task Applications in Distributed Computing Platforms

Recent emerging applications from a wide range of scientific domains often require a very large number of loosely coupled tasks to be efficiently processed. To support such applications effectively, all the available resources from different types of computing platforms such as supercomputers, grids, and clouds need to be utilized. However, exploiting heterogeneous resources from the platforms for multiple loosely coupled many-task applications is challenging, since the performance of an application can vary significantly depending on which platform is used to run it, and which applications co-run in the same node with it. In this paper, we analyze the platform and co-runner affinities of many-task applications in distributed computing platforms. We perform a comprehensive experimental study using four different platforms, and five many-task applications. We then present a two-level scheduling algorithm, which distributes the resources of different platforms to each application based on the platform affinity in the first level, and maps tasks of the applications to computing nodes based on the co-runner affinity for each platform in the second level. Finally, we evaluate the performance of our scheduling algorithm, using a trace-based simulator. Our simulation results demonstrate that our scheduling algorithm can improve the performance up to 30.0%, compared to a baseline scheduling algorithm.