Published Papers that use IRCache Traces

Wireless broadband networks are an attractive way of connecting users to the Internet. The services offered by such networks typically are point-to-point services that do not utilize the broadcast capabilities of the wireless networks. Multicast Web Caching is a readily-available and deployed method that embraces the broadcast capabilities of wireless links and largely improves web transfers. In the framework of the BROADWAN project, we have enhanced the multicast web caching solution initially developed in the EMBRACE project. Furthermore, the suitabulity of Multicast Web Caching is proven by trace-based simulation that focus on the reductions in connection latency as well as bandwith savings on both forward an return link.

In this paper we explore the evolution of both the Internet's most heavily used transport protocol, TCP, and the current network environment with respect to how the network's evolution ultimately impacts end-to-end protocols. The traditional end-to-end assumptions about the Internet are increasingly challenged by the introduction of intermediary network elements (middleboxes) that intentionally or unintentionally prevent or alter the behavior of end-to-end communications. This paper provides measurement results showing the impact of the current network environment on a number of traditional and proposed protocol mechanisms (e.g., Path MTU Discovery, Explicit Congestion Notification, etc.). In addition, we investigate the prevalence and correctness of implementations using proposed TCP algorithmic and protocol changes (e.g., selective acknowledgment-based loss recovery, congestion window growth based on byte counting, etc.). We present results of measurements taken using an active measurement framework to study web servers and a passive measurement survey of clients accessing information from our web server. We analyze our results in the context of gaining further understanding of the differences between the behavior of the Internet in theory versus the behavior we observed through measurements. In addition, these measurements can be used to guide the definition of more realistic Internet modeling scenarios.

The service time distribution is a fundamental parameter in analysis and simulation models. However, the distribution of the service times of Web servers and caches is not well understood. As a consequence, in many studies on Web servers scheduling and task assignment, the service times have been assumed as directly proportional to the size of the file transferred. But Web transfers of the same size may experience different service times due to dynamic characteristics such as server load, network traffic and client connectivity. In this paper we present models for both the service time and the file size of Web cache transfers. Our goal is to compare both empirical distributions, exploring the correlation between them. Our results are based on recent logs from NLANR. We show that both variables may be modeled by the lognormal distribution but with different parameters.

We study different delta compression policies for web access. Our emphasis is on web and proxy server-friendly policies that do not require the maintenance of multiple older versions of a page, but only use reference files accessed by the client within the last few minutes. We compare several policies for identifying appropriate reference files and evaluate their performance on a set of traces. We show that there are very simple policies that achieve significant benefits over gzip compression on most web accesses, and that can be efficiently implemented at web or proxy servers. We also investigate the potential of file synchronization techniques such as rsync for efficient web access.

Given that the global DNS system, especially at the higher root and top-levels, experiences significant query loads, we seek to answer the following questions: (1) How does the choice of DNS caching software for local resolvers affect query load at the higher levels? (2) How do DNS caching implementations spread the query load among a set of higher level DNS servers? To answer these questions we did case studies of workday DNS traffic at the University of California San Diego (USA), the University of Auckland (New Zealand), and the University of Colorado at Boulder (USA). We also tested var- ious DNS caching implementations in fully controlled laboratory experiments. This paper presents the results of our analysis of real and simulated DNS traffic. We make recommendations to network administrators and software developers aimed at improving the overall DNS system.

We propose a scalable Web document sharing infrastructure model called Browsers-Aware Proxy Server. In this design, a proxy server connecting to a group of networked clients maintains an index file of data objects of all clients' browser caches. If a user request misses in its local browser cache and the proxy cache, the browsers-aware proxy server will search the index file attempting to find it in another client's browser cache before sending the request to an upper level proxy or the Web server. If such a request does hit in a remote client, this client will directly forward the data object to the requesting client; or the proxy server fetches the data object from this client and then forwards it to the requesting client. The contributions of this caching model are twofold. First, we show that the amount of sharable data in browser caches is significant and can be utilized for document sharing among clients to improve Web caching performance and scalability. Second, the browsers-aware model can effectively and further improve Web prefetching performance. The browsers-aware model and its supported prefetching technique build a strong locality-aware Internet environment to make Web accesses fast with low communication costs. Conducting trace-driven simulations, we show the effectiveness of the browsers-aware model, and its unique advantages to facilitate prefetching.

Web content caches are often placed between end users and origin servers as a mean to reduce server load, network usage, and ultimately, user-perceived latency. Cached objects typically have associated expiration times, after which they are considered stale and must be validated with a remote server (origin or another cache) before they can be sent to a client. A considerable fraction of cache "hits" involve stale copies that turned out to be current. These validations of current objects have small message size, but nonetheless, often induce latency comparable to full-fledged cache misses. Thus, the functionality of caches as a latency-reducing mechanism highly depends not only on content availability but also on its freshness. We propose policies for caches to proactively validate selected objects as they become stale, and thus allow for more client requests to be processed locally. Our policies operate within the existing protocols and exploit natural properties of request patterns such as frequency and recency. We evaluated and compared different policies using trace-based simulations.