Shared logs offer linearizable total order across storage shards. However, they enforce this order eagerly upon ingestion, leading to high latencies. We observe that in many modern shared-log applications, while linearizable ordering is necessary, it is not required eagerly when ingesting data but only later when data is consumed. Further, readers are naturally decoupled in time from writers in these applications. Based on this insight, we propose LazyLog, a novel shared log abstraction. LazyLog lazily binds records (across shards) to linearizable global positions and enforces this before a log position can be read. Such lazy ordering enables low ingestion latencies. Given the time decoupling, LazyLog can establish the order well before reads arrive, minimizing overhead upon reads. We build two LazyLog systems that provide linearizable total order across shards. Our experiments show that LazyLog systems deliver significantly lower latencies than conventional, eager-ordering shared logs.
Do some storage interfaces enable higher performance than others? Can one identify and exploit such interfaces to realize high performance in storage systems? This article answers these questions in the affirmative by identifying nil-externality , a property of storage interfaces. A nil-externalizing (nilext) interface may modify state within a storage system but does not externalize its effects or system state immediately to the outside world. As a result, a storage system can apply nilext operations lazily, improving performance. In this article, we take advantage of nilext interfaces to build high-performance replicated storage. We implement Skyros , a nilext-aware replication protocol that offers high performance by deferring ordering and executing operations until their effects are externalized. We show that exploiting nil-externality offers significant benefit: For many workloads, Skyros provides higher performance than standard consensus-based replication. For example, Skyros offers 3× lower latency while providing the same high throughput offered by throughput-optimized Paxos.
19 August 2025 in ACM Transactions on Computer Systems
Shared logs offer linearizable total order across storage shards. However, they enforce this order eagerly upon ingestion, leading to high latencies. We observe that in many modern shared-log applications, while linearizable ordering is necessary, it is not required eagerly when ingesting data but only later when data is consumed. Further, readers are naturally decoupled in time from writers in these applications. Based on this insight, we propose LazyLog, a novel shared log abstraction. LazyLog lazily binds records (across shards) to linearizable global positions and enforces this before a log position can be read. Such lazy ordering enables low ingestion latencies. Given the time decoupling, LazyLog can establish the order well before reads arrive, minimizing overhead upon reads. We build two LazyLog systems that provide linearizable total order across shards. Our experiments show that LazyLog systems deliver significantly lower latencies than conventional, eager-ordering shared logs.
Water is an essential resource in day-to-day life. Pollution and urbanization have led to higher susceptibility of source water to contamination. There is a pressing need to develop a water quality monitoring system to preserve the quality of source water and ultimately safeguard human health. This paper proposes a low cost, wireless water quality monitoring system, wherein the quality of water stored in overhead tanks is continuously monitored. The quality of water is measured by parameters that are critical quality indicators. The data encompassing these parameters are stored in a Cloud database (in real-time) along with its timestamp. The quality of water is ascertained based on the comparison of the monitored data to standard well-established thresholds. The data, annotated with its timestamp is treated as a time-series. A univariate non-seasonal AutoRegressive Integrated Moving Average (ARIMA) model is employed to forecast individual water quality parameters. The results of forecasting is used to predict water quality deterioration. The model used is found to have mean square errors of 0.001 for pH, 0.076 for temperature and 0.001 for turbidity between the actual and forecasted values.