What the study found
The study presents Strict Timed Causal Consistency (STCC), a middleware layer for Cassandra that adds runtime-adjustable consistency with a configurable freshness bound called Δ. It reports a comparison of STCC with Cassandra’s built-in static consistency levels: ALL, QUORUM, and ONE.
Why the authors say this matters
The authors say Cassandra’s default consistency levels cannot adapt to changes in workload intensity or network conditions. They conclude that the middleware is meant to support analysis of consistency, latency, and energy trade-offs under realistic deployment conditions.
What the researchers tested
The researchers built a reproducible, open-source middleware that enforces STCC as an external consistency layer for unmodified Cassandra deployments. They tested it on a 24-node Cassandra cluster across three data centers using YCSB workloads, with CPU frequency control via cpupower and energy monitoring through SNMP-enabled PDUs and dstat.
What worked and what didn't
The abstract says the middleware integrates with unmodified Cassandra and provides tunable freshness bounds and causal guarantees across geo-replicated nodes. It also says the experiments compared STCC against ALL, QUORUM, and ONE in terms of energy, performance, and consistency trade-offs, but it does not state the detailed outcome of those comparisons in the available text.
What to keep in mind
The abstract does not provide numerical results or a clear winner among the tested configurations. It also does not describe limitations beyond the scope of the reported cluster setup and workload-based evaluation.
Key points
- The paper introduces Strict Timed Causal Consistency (STCC), a middleware layer for Cassandra.
- STCC adds a configurable freshness bound, Δ, and causal guarantees without modifying Cassandra’s core engine.
- The study compares STCC with Cassandra’s static consistency levels: ALL, QUORUM, and ONE.
- Experiments used YCSB workloads on a 24-node, three-data-center Cassandra cluster.
- The abstract mentions energy monitoring and CPU frequency control, but it does not give detailed comparative results.
Disclosure
- Research title:
- STCC middleware evaluated for Cassandra consistency trade-offs
- Authors:
- Hesam Nejati sharifaldin, Farnoush Nayebi Pour
- Institutions:
- Islamic Azad University, Mashhad
- Publication date:
- 2026-03-07
- OpenAlex record:
- View
Get the weekly research newsletter
Stay current with peer-reviewed research without reading academic papers — one filtered digest, every Friday.


