Ubiquitous mobile platforms such as Android rely on managed language run-time environments, also known as language virtual machines (VMs), to run a diverse range of user applications (apps). Each app runs in its own private VM instance, and each VM makes its own private local decisions in managing its use of processor and memory resources. Moreover, the operating system and the hardware do not communicate their low-level decisions regarding power management with the high-level app environment. This lack of coordination across layers and across apps restricts more effective global use of resources on the device.

Dynamic voltage and frequency scaling (DVFS) is ubiquitous on mobile devices as a mechanism for saving energy. Reducing the clock frequency of a processor allows a corresponding reduction in power consumption, as does turning off idle cores. Garbage collection is a canonical example of the sort of memory-bound workload that best responds to such scaling. Here, we explore the impact of frequency scaling for garbage collection in a real mobile device running Android's Dalvik virtual machine, which uses a concurrent collector. By controlling the frequency of the core on which the concurrent collector thread runs we can reduce power significantly. Running established multi-threaded benchmarks shows that total processor energy can be reduced up to 30%, with end-to-end performance loss of at most 10%.

Small mobile devices have evolved to versatile computing systems. Android devices run a complete software stack, including a full Linux kernel, user land with several software daemons and a virtual machine to run applications. On these mobile systems energy is a scarce resource and needs to be handled carefully. Current systems rely on governors that adjust the frequency of individual cores depending on the system load.

Current supercomputing systems consisting of thousands of nodes cannot meet the demands of emerging high-performance scientific applications. As a result, a new generation of supercomputing systems consisting of hundreds of thousands of nodes is being proposed. However, these systems are likely to experience far more frequent failures than today's systems, and such failures must be tackled effectively. Coordinated checkpointing is a common technique to deal with failures in supercomputers. This paper presents a model of a coordinated checkpointing protocol for large-scale supercomputers, and studies its scalability by considering both the coordination overhead and the effect of failures. Unlike most of the existing checkpointing models, the proposed model takes into account failures during checkpointing and recovery, as well as correlated failures. Stochastic activity networks (SANs) are used to model the system, and the model is simulated to study the scalability, reliability, and performance of the system.

Operator strength reduction is a technique that improves compiler-generated code by reformulating certain costly computations in terms of less expensive ones. A common case arises in array addressing expressions used in loops. The compiler can replace the sequence of multiplies generated by a direct translation of the address expression with an equivalent sequence of additions. When combined with linear function test replacement, strength reduction can speed up the execution of loops containing array references. The improvement comes from two sources: a reduction in the number of operations needed to implement the loop and the use of less costly operations.This paper presents a new algorithm for operator strength reduction, called OSR. OSR improves upon an earlier algorithm of Allen, Cocke, and Kennedy [Allen et al. 1981]. OSR operates on the static single assignment (SSA) form of a procedure [Cytron et al. 1991]. By taking advantage of the properties of SSA form, we have derived an algorithm that is simple to understand, quick to implement, and, in practice, fast to run. Its asymptotic complexity is, in the worst case, the same as the Allen, Cocke,and Kennedy algorithm (ACK). OSR achieves optimization results that are equivalent to those obtained with the ACK algorithm. OSR has been implemented in several research and production compilers.