Memory resource balancing for virtualized computing

Virtualization has become a common abstraction layer in modern data centers. By multiplexing hardware resources into multiple virtual machines (VMs) and thus enabling several operating systems to run on the same physical platform simultaneously, it can effectively reduce power consumption and building size or improve security by isolating VMs. In a virtualized system, memory resource management plays a critical role in achieving high resource utilization and performance. Insufficient memory allocation to a VM will degrade its performance dramatically. On the contrary, over-allocation causes waste of memory resources. Meanwhile, a VM’s memory demand may vary significantly. As a result, effective memory resource management calls for a dynamic memory balancer, which, ideally, can adjust memory allocation in a timely manner for each VM based on their current memory demand and thus achieve the best memory utilization and the optimal overall performance. In order to estimate the memory demand of each VM and to arbitrate possible memory resource contention, a widely proposed approach is to construct an LRU-based miss ratio curve (MRC), which provides not only the current working set size (WSS) but also the correlation between performance and the target memory allocation size. Unfortunately, the cost of constructing an MRC is nontrivial. In this dissertation, we first present a low overhead LRU-based memory demand tracking scheme, which includes three orthogonal optimizations: AVL-based LRU organization, dynamic hot set sizing and intermittent memory tracking. Our evaluation results show that, for the whole SPEC CPU 2006 benchmark suite, after applying the three optimizing techniques, the mean overhead of MRC construction is lowered from 173% to only 2%. Based on current WSS, we then predict its trend in the near future and take different strategies for different prediction results. When there is a sufficient amount of physical memory on the host, it locally balances its memory resource for the VMs. Once the local memory resource is insufficient and the memory pressure is predicted to sustain for a sufficiently long time, a relatively expensive solution, VM live migration, is used to move one or more VMs from the hot host to other host(s). Finally, for transient memory pressure, a remote cache is used to alleviate the temporary performance penalty. Our experimental results show that this design achieves 49% center-wide speedup.

Sleep Deprivation and Pain Intensity

Little or poor quality sleep is often reported in patients suffering from acute or chronic pain. Conversely, sleep loss has been known to elevate pain perception; thus a potential bi-direction relationship exists between sleep deprivation and pain. The effect of sleep deprivation on the thermal pain intensity has yet to be determined, furthermore, sex differences in pain have not been examined following sleep deprivation. There is also a higher prevalence of insomnia in women, and reports indicate that sleep quality is diminished and pain sensitivity may be greater during high hormone phases of the menstrual cycle. In Study 1 we examined the effects of 24-hour total sleep deprivation (TSD) on pain intensity during a 2-minute cold pressor test (CPT). We hypothesized that TSD would augment thermal pain intensity during CPT and women would demonstrate an elevated response compare to men. In Study 2 we investigated the effects of menstrual phase on pain intensity during CPT following TSD. We hypothesized that pain intensity would be augmented during the mid-luteal (ML) phase of the menstrual cycle. In Study 1, pain intensity was recorded during CPT in 14 men and 13 women after normal sleep (NS) and TSD. Pain intensity responses during CPT were elevated in both conditions; however, pain intensity was augmented (~ 1.2 a.u.) following TSD. When analyzed for sex differences, pain intensity was not different between men and women in either condition. In Study 2, pain intensity was recorded during CPT in 10 female subjects during the early follicular (EF) and ML phases of the menstrual cycle after TSD. Estradiol and progesterone levels were elevated during the ML phase, however, pain intensity was not different between the two phases. We conclude that TSD significantly augments pain intensity during CPT, but this response is not sex dependent. We further demonstrate that the collective effect of TSD and elevated gonadal hormone concentrations do not result in a differential pain response during the EF and ML phases of the menstrual cycle. Collectively, sleep loss augments pain intensity ratings in men and women and may contribute to sleep loss in painful conditions.