Scaling IP Lookup to Large Databases using the CRAM Lens

2025-03-04

AI Summary

To address scalability limitations in IP lookup under wide-area network growth and hardware constraints of network processors (e.g., Intel Tofino-2), this paper proposes CRAM—a unified CAM+RAM modeling framework—and introduces the first formal CRAM lens evaluation methodology, integrating TCAM-SRAM co-design principles. To overcome IPv4/IPv6 prefix lookup bottlenecks, we present three algorithms: RESAIL (for IPv4), BSIC (for IPv6), and MashUp—leveraging prefix compression, segmented indexing, and hash-enhanced techniques for algorithm-hardware co-optimization. Experiments on Tofino-2 demonstrate that RESAIL supports 2.25 million IPv4 prefixes—sufficient for projected global routing table growth over a decade—while BSIC accommodates 390K IPv6 prefixes. Both achieve throughput improvements of 9× and 3.2× over pure-TCAM baselines, respectively, significantly outperforming SAIL and Hi-BST in scalability and efficiency.

Technology Category

• Planning, Routing, and Scheduling: Routing
• Search and Optimization: Algorithm Configuration
• Machine Learning: Hardware-aware ML

Application Category

• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines
• Security and Privacy: Large-scale security measurements
• Responsible Web: Measurement, analysis, and circumvention of Web censorship

Abstract

Wide-area scaling trends require new approaches to Internet Protocol (IP) lookup, enabled by modern networking chips such as Intel Tofino, AMD Pensando, and Nvidia BlueField, which provide substantial ternary content-addressable memory (TCAM) and static random-access memory (SRAM). However, designing and evaluating scalable algorithms for these chips is challenging due to hardware-level constraints. To address this, we introduce the CRAM (CAM+RAM) lens, a framework that combines a formal model for evaluating algorithms on modern network processors with a set of optimization idioms. We demonstrate the effectiveness of CRAM by designing and evaluating three new IP lookup schemes: RESAIL, BSIC, and MashUp. RESAIL enables Tofino-2 to scale to 2.25 million IPv4 prefixes$unicode{x2014}$likely sufficient for the next decade$unicode{x2014}$while a pure TCAM approach supports only 250k prefixes, just 27% of the current global IPv4 routing table. Similarly, BSIC scales to 390k IPv6 prefixes on Tofino-2, supporting 3.2 times as many prefixes as a pure TCAM implementation. In contrast, existing state-of-the-art algorithms, SAIL for IPv4 and Hi-BST for IPv6, scale to considerably smaller sizes on Tofino-2.

Problem

• Scaling IP lookup for large databases using modern networking chips.
• Addressing hardware constraints in designing scalable IP lookup algorithms.
• Introducing CRAM lens to optimize IP lookup schemes for IPv4 and IPv6.

Innovation

• CRAM lens combines CAM and RAM for scalability
• Introduces RESAIL, BSIC, MashUp for IP lookup
• Optimizes IP lookup for modern network processors