Latencies, the time lag between the entry of input and the generation of an output from a system, are crucial to many online interactions.<\/p>\n
Although they are utilized widely and have been the subject of research for years, all existing shared logs currently suffer from high latencies. Shared logs are at the heart of many modern applications. Every cloud provider today offers a shared log service, and others provide the shared log functionality. Many hyperscalers use shared-log services for metadata.<\/p>\n
Ramnatthan Alagappan and Aishwarya Ganesan<\/p>\n
Siebel School of Computing and Data Science in The Grainger College of Engineering at the University of Illinois Urbana-Champaign assistant professors Ramnatthan Alagappan<\/a> and Aishwarya Ganesan<\/a> have received a grant from Tigerbeetle<\/a> to support their research in shared logs.<\/p>\n TigerBeetle is a financial distributed database that applies Alagappan and Ganesan\u2019s research, most notably Protocol-Aware Recovery.<\/p>\n Alagappan and Ganesan introduced protocol-aware recovery (Par), an approach that exploits protocol-specific knowledge to correctly recover from storage faults in distributed systems, in the ACM Transactions on Storage<\/a> in November 2018.<\/p>\n Alagappan presented\u00a0\u201cNew Shared-Log Abstractions for Modern Applications\u201d<\/a> on June 25, 2025, at Systems Distributed \u201825<\/a> hosted by Tigerbeetle.\u00a0Alagappan also gave a keynote, \u201cNew Shared-Log Abstractions for Modern Applications,\u201d on June 20 at the conference. The talk was broadcast on YouTube<\/a> and included a live chat.<\/p>\n Alagappan and Ganesan\u2019s research group at the University of Illinois Urbana-Champaign has been building new abstractions and designs to address the latency challenges in shared-log services. The first abstraction, LazyLog (SOSP ’24), is a new shared log better suited for applications like message queues and event-driven databases that demand low-latency ingestion. The second one, SpecLog (OSDI ’25), is a new shared-log design that reduces end-to-end latencies for critical applications like high-frequency trading, intrusion detection, and fraud monitoring.<\/p>\n Grainger Engineering Affiliations<\/p>\n