Named Data Networking (NDN) is a promising candidate of Future Internet Architecture (FIA) designed to solve many long-standing issues in current IP architecture such as security, mobility, and content distribution inefficiency. Pending Interest Table (PIT) is one of the essential building blocks of NDN, which is used in the successful delivery of data packets to the requesters. It helps to achieve intrinsic advantages such as anonymity, interest aggregation, multi-cast delivery, multi-path forwarding, and security. NFD (NDN Forwarding Daemon) currently does not constrain the size of PIT. Over-allocation of PIT does not contribute to network performance improvement rather helps in performance degradation. Therefore, there is a need to restrict PIT size. However, in the presence of bursty traffic, PIT may turn out to be full, and new incoming Interest packets may get dropped. In this case, NDN faces a new challenge to achieve Quality of Service (QoS) under bursty traffic. This paper presents two simple schemes to solve this problem via PIT entry reservation and PIT entry replacement policy. We present analytical models of both the schemes using Continuous Time Markov Chain (CTMC). We derive blocking and forced termination probability of Interest packets as performance metrics. The accuracy of the models is validated through simulation results.

命名数据网络（NDN）是未来Internet架构（FIA）的有希望的候选者，旨在解决当前IP架构中许多长期存在的问题，例如安全性，移动性和内容分发效率低下。未付利息表（PIT）是NDN的基本组成部分之一，用于成功将数据包传递给请求者。它有助于实现固有的优势，例如匿名性，兴趣聚集，多播传递，多路径转发和安全性。NFD（NDN转发守护程序）当前不限制PIT的大小。PIT的过度分配不会有助于网络性能的提高，反而会导致性能下降。因此，需要限制PIT尺寸。但是，在流量突发的情况下，PIT可能已满，并且新的传入兴趣数据包可能会被丢弃。在这种情况下，NDN面临着在突发流量下实现服务质量（QoS）的新挑战。本文提出了两种通过PIT条目保留和PIT条目替换策略解决此问题的简单方案。我们介绍使用连续时间马尔可夫链（CTMC）的两种方案的分析模型。我们导出兴趣数据包的阻塞和强制终止概率作为性能指标。通过仿真结果验证了模型的准确性。我们介绍使用连续时间马尔可夫链（CTMC）的两种方案的分析模型。我们得出兴趣包的阻塞和强制终止概率作为性能指标。通过仿真结果验证了模型的准确性。我们介绍使用连续时间马尔可夫链（CTMC）的两种方案的分析模型。我们导出兴趣数据包的阻塞和强制终止概率作为性能指标。通过仿真结果验证了模型的准确性。