DCAF：一种在线服务系统的动态计算分配框架
DCAF: A Dynamic Computation Allocation Framework for Online Serving System
摘要：
现代大型系统，如推荐系统和在线广告系统，都是建立在计算密集型基础设施之上的。在这些应用中，典型的目标是在有限的计算资源下最大化总收益，例如GMV~（总销售量）。在线服务系统通常采用多级级联结构，包括检索、预排序、排序等多个阶段，这些阶段通常采用人工分配资源的方式，要求服务配置进行相应的调整。结果，现有系统很容易陷入关于最大化总收入的次优解决方案。其局限性在于，尽管流量请求的价值变化很大，但在线服务系统仍然在它们之间花费相等的计算能力。在本文中，我们介绍了一种新的思想，在线服务系统可以对待每个流量请求不同，并分配“个性化”的计算资源的基础上，其价值。我们将这个资源分配问题转化为背包问题，并提出了一个动态计算分配框架DCAF。在一定的假设条件下，DCAF理论上可以保证系统在给定的计算预算下，使总收益最大化。DCAF带来了显著的改进，并已部署在淘宝的展示广告系统中，为主要流量服务。通过DCAF，我们能够在减少20%计算资源的情况下保持相同的业务性能。
英文摘要：
Modern large-scale systems such as recommender system and online advertising system are built upon computation-intensive infrastructure. The typical objective in these applications is to maximize the total revenue, e.g. GMV~(Gross Merchandise Volume), under a limited computation resource. Usually, the online serving system follows a multi-stage cascade architecture, which consists of several stages including retrieval, pre-ranking, ranking, etc. These stages usually allocate resource manually with specific computing power budgets, which requires the serving configuration to adapt accordingly. As a result, the existing system easily falls into suboptimal solutions with respect to maximizing the total revenue. The limitation is due to the face that, although the value of traffic requests vary greatly, online serving system still spends equal computing power among them. In this paper, we introduce a novel idea that online serving system could treat each traffic request differently and allocate "personalized" computation resource based on its value. We formulate this resource allocation problem as a knapsack problem and propose a Dynamic Computation Allocation Framework~(DCAF). Under some general assumptions, DCAF can theoretically guarantee that the system can maximize the total revenue within given computation budget. DCAF brings significant improvement and has been deployed in the display advertising system of Taobao for serving the main traffic. With DCAF, we are able to maintain the same business performance with 20\% computation resource reduction.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
