Yongjin Choi

Mirinae: A peer-to-peer overlay network for large-scale content-based publish/subscribe systems

Content-based publish/subscribe systems provide a useful alternative to traditional address-based communication due to their ability to decouple communication between participants. It has remained a challenge to design a scalable overlay supporting the complexity of content-based networks, while satisfying the desirable properties large distributed systems should have. This paper presents the design of Mirinae, a new structured peer-to-peer overlay mesh based on the interests of peers. Given an event, Mirinae provides a flexible and efficient dissemination tree minimizing the participation of non-matching nodes. We also present a novel ID space transformation mechanism for balancing routing load of peers even with highly skewed data, which is typical of the real world. Mirinae can be used as a substrate for content-search and range query in other important distributed applications.

DFS: a de-fragmented file system

Small file accesses are still limited by disk head movement on modern disk drives with the high disk bandwidth. Small file performance can be improved by grouping and clustering, each of which places multiple files in a directory and places blocks of the same file on disks contiguously. These schemes make it possible for file systems to use large data transfers in accessing small files, reducing disk accesses. However, as file systems become aged, disks become too fragmented to support the grouping and clustering of small files. This fragmentation makes it difficult for file systems to take advantage of large data transfers, increasing disk I/Os. To offer a solution to this problem, we describe a de-fragmented file system (DFS). By using data cached in memory, DFS relocates and clusters data blocks of small fragmented files in a dynamic manner. Besides, DFS clusters related small files in the same directory, at contiguous disk locations. Measurements of DFS implementation show that the techniques alleviate file fragmentation significantly and, in particular performance for small file reads exceeds that of a traditional file system by 78%.

Associativity based clustering and query stride for on-demand routing protocols in ad hoc networks

An ad hoc network is a collection of wireless mobile nodes without the support of a stationary infrastructure. In ad hoc networks, each node acts as a router to support multiple hops to overcome limited range of packet radios. While a variety of routing protocols have been developed recently, a class of routing schemes called on-demand protocols have attracted a lot of attentions due to their low routing overhead. However, their efficiency is limited by the enormous query flooding overhead and the route acquisition latency. In this paper, we propose two new schemes — Associativity Based Clustering (ABC) and Query Stride (QS) — which offer benefits to on-demand protocols by surmounting those limitations. First, ABC provides a proactive routing information on associatively-stable nodes in the network. Thus, nodes can communicate immediately with a node in ABC without any query-reply exchange. Second, QS enables a query to be flooded to the whole network with a minimum number of broadcasts. While ABC reduces the route acquisition latency, QS saves the bandwidth in flooding. Although ABC and QS can be used independently, the combined effect is even greater since they provide complementary benefits sharing common overhead.

Mirinae: A Peer-to-Peer Overlay Network for Content-Based Publish/Subscribe Systems

Content-based publish/subscribe systems provide a useful alternative to traditional address-based communication due to their ability to decouple communication between participants. It has remained a challenge to design a scalable overlay supporting the complexity of content-based networks, while satisfying the desirable properties large distributed systems should have. This paper presents the design of Mirinae, a new structured peer-to-peer overlay mesh based on the interests of peers. Given an event, Mirinae provides a flexible and efficient dissemination tree minimizing the participation of non-matching nodes. We also present a novel ID space transformation mechanism for balancing routing load of peers even with highly skewed data, which is typical of the real world. Our evaluation demonstrates that Mirinae is able to achieve its goals of scalability, efficiency, and near-uniform load balancing. Mirinae can be used as a substrate for content-search and range query in other important distributed applications.

Network Iron Curtain: Hide Enterprise Networks with OpenFlow

In this paper, we propose a new network architecture, Network Iron Curtain that can handle network scanning attacks automatically. Network Iron Curtain does not require additional devices or complicated configurations when it detects scanning attack, and it can confuse scanning attackers by providing fake scanning results. When an attacker sends a scanning packet to a host in Network Iron Curtain, Network Iron Curtain detects this trial and redirects this packet to a honeynet, which is installed with Network Iron Curtain. The honeynet will respond to this scanning packet based on the predefined policy instead of the original target host. Therefore, the attacker will have fake information (i.e., false open port information). We implement a prototype system to verify the proposed architecture, and we show an example case of detecting network scanning.

S-VFS: Searchable Virtual File System for an Intelligent Ubiquitous Storage

With advances in ubiquitous environments, user demand for easy data-lookup is growing rapidly. Not only users but intelligent ubiquitous applications also require data-lookup services for a ubiquitous computing framework. This paper proposes a backward-compatible, searchable virtual file system (S-VFS) for easy data-lookup. We add search functionality to the VFS, the de facto standard abstraction layer over the file system. Users can find a file by its attributes without remembering the full path. S-VFS maintains the attributes and the indexing structures in a normal file per partition. It processes queries and returns the results in a form of a virtual directory. S-VFS is the modified VFS, but uses legacy file systems without any modification. Since S-VFS supports full backward compatibility, users can even browse hierarchically with the legacy path name. We implement S-VFS in Linux kernel 2.6.7-21. Experiments with randomly generated queries demonstrate outstanding lookup performance with a small overhead for indexing.

MTCP: A Transmission Control Protocol for Multi-Provider Environment

Transmission schemes that gain content from multiple servers concurrently have been highlighted due to their ability to provide bandwidth aggregation, stability on dynamic server departure, and load balancing. Previous approaches employ parallel downloading in the transport layer to minimize the receiver buffer size and maximize bandwidth utilization. However, they only focus on the receiver operations and induce considerable overhead at the senders in contradiction to the main goal of a multi-provider environment, offloading popular servers through replication. In the present work, the authors propose MTCP, a novel transport layer protocol that focuses on reduction of the sender overhead through the elimination of unnecessary disk I/Os and efficient buffer cache utilization. MTCP also balances trade-off objectives to minimize buffering at receivers and maximize the request locality at senders.

New peer-to-peer overlay network for content-based publish/subscribe systems

Publish/subscribe has become a prevalent model for distributed content delivery due to its ability to decouple communication between participants. It has remained a challenge to design a scalable overlay supporting the complexity of content-based publish/subscribe systems, while satisfying the desirable properties large distributed systems should have. This paper presents the design of Mirinae, a new structured peer-to-peer overlay mesh based on the interests of peers. To fulfil the gap between the rich subscription language of publish/subscribe and the logical address of overlay network, we devise a bloom-filter based mapping scheme. Given an event, Mirinae provides a flexible and efficient dissemination tree minimizing the participation of non-matching nodes. Our evaluation demonstrates that Mirinae is able to achieve its goals of scalability, efficiency, and adaptiveness.