Future 5G mobile networks are expected to support high numbers of simultaneously connected devices and to achieve high system spectrum efficiency at low power consumption. To attain these goals, device-to-device (D2D) communication is a key component. We allow any cellular user equipment to share its radio resource with multiple D2D devices and then develop the MiSo algorithm, which considers spectrum reuse and power control jointly. Simulation results confirm that compared with several algorithms, MiSo achieves highest system throughput and power efficiency with shortest run time.

The power-saving opportunities of long-running application servers which execute on multi-core systems are studied in this paper. The research goal is to develop an efficient power-saving strategy of application servers with the minimum performance degradation in cloud environments. The power-saving strategy is based on the run-time information which is already available in a JVM, the base software component of application servers. Several key findings are revealed through this study. First, the particular behavior of application servers, also known as phases, can be related to the run-time information of a JVM. Thus the phases of an application server can be predicted before the applications actually execute on hardware. Secondly, some particular phases are observed in this study and used to establish the power-saving strategy, such as memory phases and execute phases. Finally, a new finding of idle phase is proposed to reduce significant energy wastage without performance degradations. Based on these findings, a set of power-saving algorithms is proposed and implemented with two widely used JVMs, Sun's Hotspot and Jikes RVM. With the experiments of five multi-threaded benchmarks and two web application benchmarks, the use of proposed power-saving strategy leads to the lowest value of energy-delay product among the other power-saving techniques, and the performance degradation is well below 6 percent.

Compared to current mobile networks, next-generation mobile networks are expected to support higher numbers of simultaneously connected devices and to achieve higher system spectrum efficiency and lower power consumption. To achieve these goals, we study the multi-sharing device-to-device (D2D) communication, which allows any cellular user equipment to share its radio resource with multiple D2D devices. We jointly consider resource block reuse and power control and then develop the MISS algorithm. Simulation results show that MISS performs very well in terms of transmission power consumption, system throughput, and the number of permitted D2D devices.

Device-to-device (D2D) communication can improve system spectrum efficiency in mobile networks; however, existing D2D resource allocation techniques may not work well in 5G mobile networks with many simultaneously connected devices. To address this problem, we study the multi-sharing resource allocation problem, which allows any cellular user equipment to share its radio resource with multiple D2D devices. We formulate the multi-sharing resource allocation problem and prove its NP hardness. Besides, we develop the Greedy Throughput Maximization Plus (GTM+) algorithm. Simulation results show that GTM+ is fast and outperforms existing algorithms in throughput and the number of permitted D2D pairs.

Multi-core processors are becoming widely deployed in computer systems. While improving the performance of applications with multi-core technology, multi-core CPUs consume more power than a single-core one. We propose a power saving technique for multi-core systems based on the observation of critical sections in multi-threaded applications. First, since only one thread on a core can enter into a critical section for a shared resource, other threads on other cores that would access the same resource would run into busy waiting state. Hence the frequency of those idle cores could be minimized to reduce energy wastages without degrading overall applications¡¦ performance. Secondly, the accurate engage/disengage timing of a critical section could be detected by matching particular byte code patterns before a thread actually enters into a critical section. Thus the frequency of busy waiting cores could be tuned in advance. Based on these findings, the power-saving technique of critical sections is proposed and implemented. With the experiments of five multi-threaded Java benchmarks, the proposed power-saving technique leads to good energy savings (11 to 15 percent) with lower values of Energy Delay Product as compared to the other power-saving techniques.

The secure requirement of modern information systems is increasing significantly nowadays, especially in cloud computing with distributed applications. Among secure mechanisms of an organization, the access control (AC) is a foundation for modern information systems. In order to reach an effective and flexible approach of access control, the Role-based Access Control (RBAC) models are widely deployed in organizations. Comparing with traditional discretionary access control (DAC) and mandatory access control (MAC), the user-role-subject mapping of a RBAC model provides effective permissions assignments for access control of an organization. However, the RBAC sessions could be stretched over multiple distributed servers in cloud computing. The traversed sessions between servers could be modified, monitored and attacked by modern hacker techniques, and become secure leaks in RBAC models. In order to ensure secure sessions in cloud computing, various encryption approaches are used. Among these encryption approaches, the use of digital certificates by asymmetric encryption methods could be an appropriate solution to ensure the security of RBAC sessions. However, the cost of public/private keys management and issuing an appropriate certificate tokens for each member in organizations could be very expensive. The high cost might block the deployment of secure RBAC sessions, and then reduce the secure level of organizations. In order to improve this issue, a low-cost approach of secure sessions for RBAC models is proposed in this paper. The personal smart cards can be used as a certificate tokens in RBAC models to reach effective user authentications. Moreover, each session of RBAC models, including user-role-subject assignments and content-based accesses, can be protected by digital certificates which is generated by user own smart cards. Thus the security of RBAC sessions can be improved significantly. It is worth noting that personal smart cards are issued by public departments, thus the expense of tokens issuing and key management could be minimized. Therefore, the session security of a RBAC model could be ensured with user own smart cards without additional cost.

The performance and scalability issues of multithreaded Java programs on multicore systems are studied in this paper. First, we examine the performance scaling of benchmarks with various numbers of processor cores and application threads. Second, by correlating low-level hardware performance data to JVM threads and system components, the detail analyses of performance and scalability are presented, such as the hardware stall events and memory system latencies. Third, the usages of memory resource are detailed to observe the potential bottlenecks. Finally, the JVM tuning techniques are proposed to alleviate the bottlenecks, and improve the performance and scalability. Several key findings are revealed through this study. First, the lock contentions usually lead to a strong limitation of scalability. Second, in terms of memory access latencies, the most of memory stalls are produced by L2 cache misses and cache-to-cache transfers. Finally, the overhead of minor garbage collections could be an important factor of throughput reductions. Based on these findings, the appropriate Java Virtual Machine (JVM) tuning techniques are examined in this study. We observe that the use of a parallel garbage collector and an appropriate ratio of young to old generation can alleviate the overhead of minor collection and improve the efficiency of garbage collections. Moreover, the cache utilizations could be enhanced with the use of thread-local allocation buffer, and then leads to the performance improvements significantly.