Abstract-In this paper, we propose a two-phase methodology for systematically evaluating the performability (performance and availability) of cluster-based Internet services. In the first phase, evaluators use a fault-injection infrastructure to characterize the service's behavior in the presence of faults. In the second phase, evaluators use an analytical model to combine an expected fault load with measurements from the first phase to assess the service's performability. Using this model, evaluators can study the service's sensitivity to different design decisions, fault rates, and other environmental factors. To demonstrate our methodology, we study the performability of a multitier Internet service. In particular, we evaluate the performance and availability of three soft state maintenance strategies for an online bookstore service in the presence of seven classes of faults. Among other interesting results, we clearly isolate the effect of different faults, showing that the tier of Web servers is responsible for an often dominant fraction of the service unavailability. Our results also demonstrate that storing the soft state in a database achieves better performability than storing it in main memory (even when the state is efficiently replicated) when we weight performance and availability equally. Based on our results, we conclude that service designers may want an unbalanced system in which they heavily load highly available components and leave more spare capacity for components that are likely to fail more often.
-e-Commerce services have become a promising and profitable application of the Internet. In order to keep them growing, solutions must be found to deal with unreliable connections and high latencies, among other problems. The best solutions to such problems tend to depend on the distribution of the service over the network, placing servers in multiple locations, closer to customers. If placement of servers is effective it tends to reduce delays and traffic-related costs. In this paper we discuss the distribution of e-commerce services by introducing a traffic-aware cost model and evaluating it using an actual log from an e-tailer. The results show that the model yields good placement solutions, which perform better than simpler ad-hoc solutions.I INTRODUCTION e-Commerce services have become a promising and profitable application of the Internet. Nevertheless, the use of the Internet as the communication environment in this case is not without its problems. For an e-commerce site to be successful it has to reduce the negative effects of unreliable connections and high network latencies, among other problems. Such problems may compromise customer satisfaction and therefore the success of the virtual business.Techniques based on improving the quality of a centralized server do not provide a reliable solution, since the difficulties faced are inherent to the network infrastructure itself, not just the server. A better solution tends to be the distribution of the service over the network, placing servers in multiple locations closer to the final users. The existence of multiple servers tends to increase availability, and assuming the placement is well planned, it tends to reduce delays. Proxy cache servers [8] are an example of a successful service distribution strategy: they replicate static content such as HTML pages and images, and are responsible for significant traffic reduction. Content distribution networks [3] have deserved special attention lately, since they provide a more reliable and controlled solution than proxy caches, but at a cost for the information providers. On the other hand, there are several applications that generate dynamic and non-cacheable responses (at least when using traditional cache technologies).For a distributed e-commerce solution to be successful, there are several issues that must be addressed, such as resource placement and discovery, and load balancing. Resource placement policies define the number and location of servers and products. A larger number of servers allows a higher degree of parallelism, but also leads to higher complexity and context maintenance costs. Regarding product allocation in servers, it would be simpler to divide them uniformly, but a strategy that takes regional demands into account should be more cost-effective. Resource discovery is defined by strategies that determine how clients decide which server to contact. These strategies may be based on network topology or geographical affinity, and may be implemented through dynamic DNS replies, for ex...
e-Commerce services have become a promising and pro table application of the Internet. Nevertheless, the use of the Internet as the communication environment i n this case is not without its problems. For an e-commerce site to be successful it has to reduce the negative e ects of unreliable connections and high network latencies, among others. Such problems may compromise customer satisfaction and therefore the success of the virtual business. Techniques based on improving the quality of a centralized server do not provide a reliable solution, since the di culties faced are inherent to the network infrastructure itself, not just the server. A better solution tends to be the distribution of the service over the network, placing servers in multiple locations closer to the nal users. The existence of multiple servers tends to increase availability, and assuming the placement i s w ell planned, it tends to reduce delays and tra c-related costs. In this paperwe discuss the distribution of e-commerce services by i n troducing a tra c-aware cost model and evaluating it through an actual log from an e-tailer. The results show that the model allow system designers to investigate cost compromises and the impact of the application workload on the e ectiveness of the distribution in terms of reducing tra c, and thus operational costs and user-perceived latency.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.