Search for: All records

Memory allocation is increasingly important to parallel performance, yet it is challenging because a program has data of many sizes, and the demand differs from thread to thread. Modern allocators use highly tuned heuristics but do not provide uniformly good performance when the level of concurrency increases from a few threads to hundreds of threads. This paper presents a new timescale theory to model the memory demand in real time. Using the new theory, an allocator can ad- just its synchronization frequency using a single parameter called allocations per fetch (apf ). The paper presents the timescale the- ory, the design and implementation of APF tuning in an existing allocator, and evaluation of the effect on program speed and mem- ory efficiency. APF tuning improves the throughput of MongoDB by 55%, reduces the tail latency of a Web server by over 60%, and increases the speed of a selection of synthetic benchmarks by up to 24× while using the same amount of memory. 

Caching techniques are widely used in today’s computing infrastructure from virtual memory management to server cache and memory cache. This paper builds on two observa- tions. First, the space utilization in cache can be improved by varying the cache size based on dynamic application demand. Second, it is easier to predict application behavior statistically than precisely. This paper presents a new variable-size cache that uses statistical knowledge of program behavior to maximize the cache performance. We measure performance using data access traces from real-world workloads, including Memcached traces from Facebook and storage traces from Microsoft Research. In an offline setting, the new cache is demonstrated to outperform even OPT, the optimal fixed- size cache which makes use of precise knowledge of program behavior. 

Serotonin neurons of the dorsal and median raphe nuclei (DR, MR) collectively innervate the entire forebrain and midbrain, modulating diverse physiology and behavior. To gain a fundamental understanding of their molecular heterogeneity, we used plate-based single-cell RNA-sequencing to generate a comprehensive dataset comprising eleven transcriptomically distinct serotonin neuron clusters. Systematic in situ hybridization mapped specific clusters to the principal DR, caudal DR, or MR. These transcriptomic clusters differentially express a rich repertoire of neuropeptides, receptors, ion channels, and transcription factors. We generated novel intersectional viral-genetic tools to access specific subpopulations. Whole-brain axonal projection mapping revealed that DR serotonin neurons co-expressing vesicular glutamate transporter-3 preferentially innervate the cortex, whereas those co-expressing thyrotropin-releasing hormone innervate subcortical regions in particular the hypothalamus. Reconstruction of 50 individual DR serotonin neurons revealed diverse and segregated axonal projection patterns at the single-cell level. Together, these results provide a molecular foundation of the heterogenous serotonin neuronal phenotypes.