Latency-critical applications directly interact with end users and often experience the diurnal load pattern. In production, best-effort applications are often co-located with them to utilize the idle cores at the low load. Meanwhile, modern computers are evolving towards heterogeneous NUMA architecture, where the cores have different computation abilities, memory access latencies and network communication delays. Prior co-location scheduling work did not consider the NUMA architecture, and failed to maximize the throughput of best-effort applications while ensuring the required QoS of latency-critical applications. Our investigation shows that NUMA effect has complex impacts on the latency of latency-critical applications and the throughput of best-effort applications. We therefore propose PAC, a preference-aware co-location scheduling scheme that considers the NUMA effect for heterogeneous NUMA architectures. PAC has a performance monitor and a core scheduler. Specifically, the performance monitor identifies the "dangerous" latency-critical applications that require upgrading core allocations. We propose two low-overhead scheduling strategies for the scheduler. The strategies identify the bottlenecks of applications and adjust core allocations accordingly. Experimental result shows that PAC improves the throughput of best-effort applications by 3.87× while ensuring the required QoS of latency-critical applications.
Emerging evidence has suggested that the pathophysiology of obsessive compulsive disorder (OCD) might involve widely distributed large-scale brain systems. The dysfunction within salience network, which is comprised of dorsal anterior cingulated cortex (dACC) and bilateral insular areas, has been proposed to contribute to OCD onset. The mechanism underlying salience network abnormality remains unclear and it is worthwhile to investigate its clinical relevance using functional neuroimaging approaches. In this study, we performed the spontaneous brain activity measurement using resting-state functional magnetic resonance imaging (fMRI) on unmedicated OCD patients (n=23). Specifically, the amplitude of low frequency (0.01-0.08 Hz) fluctuations (ALFF) was calculated for regions in salience network. The voxel-based Pearson's correlative analysis was conducted to explore the relationship beween ALFF measures and symptom severity for OCD patients. The results showed that the spontaneous neuronal activity in insula was significantly correlated to OCD clinical symptoms, especially compulsive behaviors. Our findings consolidated that the salience network played an important role in the pathogenesis of OCD and the intensity of intrinsic brain activity in insula provided a predictive biomarker for OCD symptom severity.