Theses and Dissertations from UMD
Permanent URI for this communityhttp://hdl.handle.net/1903/2
New submissions to the thesis/dissertation collections are added automatically as they are received from the Graduate School. Currently, the Graduate School deposits all theses and dissertations from a given semester after the official graduation date. This means that there may be up to a 4 month delay in the appearance of a give thesis/dissertation in DRUM
More information is available at Theses and Dissertations at University of Maryland Libraries.
Browse
3 results
Search Results
Item PARALLEL COMPUTING WITH P2P DESKTOP GRIDS(2015) Jackson, Gary Lee; Sussman, Alan; Computer Science; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Tightly-coupled parallel computing is an important tool for problem solving. Structured peer-to-peer network overlays are failure-tolerant and have a low admin- istrative burden. This work seeks to unite the two. First, I present a completely decentralized algorithm for parallel job scheduling and load balancing in distributed peer-to-peer environments. This algorithm is useful for meta-scheduling across known clusters and scheduling on desktop grids. To accomplish this, I build on previous work to route jobs to appropriate resources then use the new algorithm to start parallel jobs and balance load across the grid. I also discuss what constitutes useful clusterings for this algorithm as well as inherent scaling limitations. Ultimately, I show that my algorithm performs comparably to one using centralized load balancing with global up-to-date information. The principal contribution of this work is that the parallel job scheduling is completely decentralized, which is not featured in previous work, and enables reliable ad hoc sharing of distributed resources to run parallel computations. Second, I show how clusters of computers can be found dynamically by using an existing latency prediction technique coupled with a new refinement algorithm. Several latency prediction techniques are compared experimentally. One, based on a tree metric space embedding, is found to be superior to the others. Nevertheless, I show that it is not quite accurate enough. To solve this problem, I present a refinement algorithm for producing quality clusters while still maintaining bounds for the amount of information any given node must store about other nodes. I show that clusters derived this way have scheduler performance comparable to those chosen statically with global knowledge. Lastly, I discuss previously undiscovered under-specifications in the Content Addressable Network (CAN) structured peer to peer system. In high-churn situ- ations, the CAN allows stale information and changes to the overlay structure to create routing problems. I show solutions to these two problems, as well as discuss other issues that may also disrupt a CAN.Item Fast Scalable Peer-to-Peer Lookup Services for Multi-Hop Wireless Networks(2008-01-28) Shin, Min-Ho; Arbaugh, William A; Computer Science; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Recent years have seen growing popularity of multi-hop wireless networks such as wireless mesh networks and sensor networks. Such systems require efficient lookup services for reliable system operation such as packet routing, key-discovery, and object lookup. The lack of infrastructure, however, makes the centralized lookup fail to scale in multi-hop wireless networks. For example, consider a citywide wireless mesh network which provides wireless connection service to a number of mobile users. Due to a high volume of user access and inherent vulnerability of wireless links, centralized authentication methods fail to scale. The decentralization of user authentication, however, faces a challenge of key discovery ; how to find the location of user keys. Motivated from the user authentication problem in wireless mesh networks, this dissertation work aims to provide efficient and scalable distributed lookup services for multi-hop wireless networks. Employing the notion of peer-to-peer lookup where each node can both query and respond, I present two different methods: Valley-Walk and Rigs. A loosely-structured scheme Valley-Walk strategically places object copies and locates them efficiently only with a minimal local structure. The Valley-Walk finds target objects in near-optimal hop counts with a moderate number of copies (e.g., 10% the network size) stored in the network. Without a global structure, however, Valley-Walk fails to guarantee the low cost search with a small number of copies. A tightly-structured scheme Rigs (Ring Interval Graph Search) realizes a Distributed Hash Table (DHT) in multi-hop wireless networks. Experimental study shows the limitations of existing DHTs in mult-hop wireless networks due to its independence of underlying topology. Unlike DHT, Rigs constructs a search structure Ring Interval Graph such that queries are forwarded only to local neighbors. Rigs guarantees successful object lookup with near-optimal performance.Item Lookup Protocols and Techniques for Anonymity(2006-08-18) Morselli, Ruggero; Katz, Jonathan; Bhattacharjee, Bobby; Computer Science; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)This dissertation covers two topics of interest for network applications: lookup protocols, a basic building block for distributed systems, and ring signatures, a powerful primitive for anonymous communication. In the first part of this work, we review lookup protocols, distributed algorithms that allow users to publish a document as well as to look up a published document that matches a given name. Our first major contribution is to design Local Minima Search (LMS), a new efficient lookup protocol for a model in which a node is physically connected to a few other nodes and may only communicate directly with them. Our second major contribution is the formulation of a new model in which we allow an arbitrary number of misbehaving nodes, but we assume a restriction on their network addresses. We then design a new lookup protocol for this setting. In the second part of this dissertation, we present our work on ring signatures, a variant of digital signatures, which enables a user to sign a message so that a set of possible signers is identified, without revealing which member of that set actually generated the signature. Our first contribution on this topic is new definitions of security which address attacks not taken into account by previous work. As our second contribution, we design the first provably secure ring signature schemes in the standard model.