Latency-critical Applications Research Articles

Agile C-states: A Core C-state Architecture for Latency Critical Applications Optimizing both Transition and Cold-Start Latency

Latency critical applications running in modern datacenters exhibit irregular request arrival patterns and are implemented using multiple services with strict latency requirements (30 μs –250 μs ). These characteristics render existing energy saving idle CPU sleep states ineffective due to the performance overhead caused by the state’s transition latency. Besides the state transition latency, another important contributor to the performance overhead of sleep states is the cold-start latency, or in other words, the time required to warm-up microarchitectural state (e.g., cache contents, branch predictor metadata) that is flushed or discarded when transitioning to a lower-power state. Both the transition latency and cold-start latency can be particularly detrimental to the performance of latency critical applications with short execution times. While prior work focuses on mitigating the effects of transition and cold-start latency by optimizing request scheduling, in this work, we propose a redesign of the Core C-state architecture for latency-critical applications. In particular, we introduce C6Awarm a new Agile Core C-state that drastically reduces the performance overhead caused by idle sleep state transition latency and cold-start latency, while maintaining significant energy savings. C6Awarm achieves its goals by implementing 1) medium-grained power gating, 2) preserving the microarchitectural state of the core and 3) by keeping the clock generator and PLL active and locked. Our analysis for a set of microservices based on an Intel Skylake server, shows that C6Awarm manages to reduce the energy consumption by up to \(70\% \) with limited performance degradation (at-most \(2\% \) ).

ISwap: A New Memory Page Swap Mechanism for Reducing Ineffective I/O Operations in Cloud Environments

This paper proposes iSwap, a new memory page swap mechanism that reduces the ineffective I/O swap operations and improves the QoS for applications with a high priority in the cloud environments. iSwap works in the OS kernel. iSwap accurately learns the reuse patterns for memory pages and makes the swap decisions accordingly to avoid ineffective operations. In the cases where memory pressure is high, iSwap compresses pages that belong to the latency-critical (LC) applications (or high-priority applications) and keeps them in main memory, avoiding I/O operations for these LC applications to ensure QoS; and iSwap evicts low-priority applications’ pages out of main memory. iSwap has a low overhead and works well for cloud applications with large memory footprints. We evaluate iSwap on Intel x86 and ARM platforms. The experimental results show that iSwap can significantly reduce ineffective swap operations (8.0% - 19.2%) and improve the QoS for LC applications (36.8% - 91.3%) in cases where memory pressure is high, compared with the latest LRU-based approach widely used in modern OSes.

Prebaking runtime environments to improve the FaaS cold start latency

Function-as-service (FaaS) platforms promise a simpler programming model for cloud computing, given that providers take care of the overall resource management while the developers can concentrate only on writing their applications in the scope of a function, not having to care about installing or scaling resources. As FaaS users are billed based on the execution of the functions, platform providers have a natural incentive not to keep idle resources running at the platform’s expense. However, this strategy may lead to the cold start issue, in which the execution of a function is delayed because there are no ready resources to host the execution. Despite the time to provision the computational resources, starting the runtime environment that manages the function execution can take hundreds of milliseconds to seconds, a prohibitive non-deterministic overhead to latency-critical applications. This work describes a practical technique, Prebaking, to reduce the function start-up time based on restoring snapshots of previously executed processes. The foundation of the Prebaking technique is that deciding when to create a snapshot of a function is essential. For example, based on a prototype we developed using the CRIU checkpoint/restore Linux tool, the performance improvement on the start-up of a properly snapshotted function is up to 25 times. Fortunately, generating a good function snapshot is easy: One only needs to checkpoint a warm function, that is, a function that has processed a request. To show the feasibility of using the Prebaking technique, we discussed integrating it into OpenFaaS, an open-source FaaS platform, including how to generate and restore function snapshots. We also evaluated our prototype by running a comprehensive set of experiments that compare the function start-up duration and cold start latency against the standard fork-exec procedure. We analyze the JVM, CPython, and Node.js runtimes. The results indicate that the technique can improve the function replica start-up time even for no-warmed functions: we improved the function replica start-up time from 3.5 times for a “do-nothing” function running in Node.js up to 12 times when considering an Image Resizer function running in CPython. Finally, we reinforce the practicability of the proposed technique by comparing our proposal to another snapshot-based technique called SEUSS. That evaluation shows that Prebaking provides better cold start latency even for slightly complex functions like the Markdown Renderer.

Metaheuristics Method for Computation Offloading In Mobile Edge Computing: Survey

In recent years, edge computing has emerged as a computing paradigm to support the computationally intensive and latency-critical applications for resource limited devices. The main feature of edge computing is to push computation, networking, and storage facilities closer to the network edge. This enables user equipment (UE) to profit from the edge computing paradigm by mainly offloading their intensive computation tasks to edge resources. Because of some rising problems such as inherent software and hardware heterogeneity, restrictions, dynamism, and stochastic behavior of the ecosystem, the computation offloading issues consider as the essential challenging problems in the MEC environment. However, to the best of the author's knowledge, in spite of its significance, in Metaheuristic -based computation offloading mechanisms, there is not any systematic, comprehensive, and detailed survey in the MEC environment. In this paper, we provide a review on the Metaheuristic -based computation offloading mechanisms in the MEC environment in the form of a classical taxonomy to identify the contemporary mechanisms on this crucial topic and to offer open issues as well. The proposed taxonomy is classified into three main fields: genetic algorithm (GA)-based mechanisms, Partical Swarm Optimization(PSO)-based mechanisms, and Hybrid GA with PSO -based mechanisms. Next, these classes are compared with each other based on the essential features such as performance metrics, case studies, utilized techniques, and evaluation tools, and their advantages and weaknesses are discussed, as well. Finally, open issues and uncovered or inadequately covered future research challenges are argued, and the survey is concluded.

Efficient Mobility Management in Mobile Edge Computing Networks: Joint Handover and Service Migration

Mobile edge computing (MEC) has been envisioned as an essential technology for latency-critical applications by providing computing services in close proximity to mobile users. Bringing MEC to come into practice, how to support user mobility remains challenging. In addition to seeking a thorny trade-off between service latency and migration cost, both interactions in space and time exist in mobility management, which requires collaboration among users and perfect prior knowledge, including user mobility and network information. In this paper, we propose an efficient mobility management framework for MEC networks, in which mobility management is operated centered around users’ performance and cost, while radio access and computing service provision are loosely coupled. With a loosely coupled design, the proposed framework exhibits more flexibility and incurs higher complexity. Focusing on multi-user and multi-cell MEC networks, this joint control problem is formulated to maximize the long-term total utility accounting for the service delay and migration cost. Considering the exponential complexity, a distributed mobility management approach is developed, which combines game theory and user-oriented deep reinforcement learning to deal with the interactions in space and time. Simulation results show the efficiency and scalability of the proposed approach.

Cooperative Service Placement and Request Routing in Mobile Edge Networks for Latency-Sensitive Applications

The emergence of mobile applications like self-driving cars, networked gaming, virtual reality, and augmented reality has created a demand for computation-intensive and latency critical applications. The dense deployment of base stations (BSs) makes the overlapping coverage of BS that induces the placement of services and routing the requests to the appropriate node is a challenging issue. Further, the limited resources at the edge node invite additional problems of allocating the existing limited edge resources cooperatively to fulfil the user demands. To address the problems mentioned above, we formulate the cooperative service placement and routing problem by placing the diversified latency critical services to maximize the time utility with the deadline and resource constraints. The problem is formulated as an integer linear programming problem for cooperative service placement in mobile edge networks. A greedy and randomized rounding technique is presented to solve the service caching and request routing problems decomposed from the proposed problem. Further, a heuristic algorithm based on the circular convex set is presented to solve the proposed NP-hard problem. Extensive simulation results demonstrate that the proposed cooperative service placement schemes significantly improve the delay, cache hit ratio, load on cloud, and time utility performance compared with existing mechanisms.

A Markov Decision Process Solution for Energy-Saving Network Selection and Computation Offloading in Vehicular Networks

Vehicular Edge Computing (VEC) enables the integration of edge computing facilities in vehicular networks (VNs), allowing data-intensive and latency-critical applications and services to end-users. Though VEC brings several benefits in terms of reduced task computation time, energy consumption, backhaul link congestion, and data security risks, VEC servers are often resource-constrained. Therefore, the selection of proper edge nodes and the amount of data to be offloaded becomes important for having VEC process benefits. However, with the involvement of high mobility vehicles and dynamically changing vehicular environments, proper VEC node selection and data offloading can be challenging. In this work, we consider a joint network selection and computation offloading problem over a VEC environment for minimizing the overall latency and energy consumption during vehicular task processing, considering both user and infrastructure side energy-saving mechanisms. We have modeled the problem as a sequential decision-making problem and incorporated it in a Markov Decision Process (MDP). Numerous vehicular scenarios are considered based upon the users' positions, the states of the surrounding environment, and the available resources for creating a better environment model for the MDP analysis. We use a value iteration algorithm for finding an optimal policy of the MDPs over an uncertain vehicular environment. Simulation results show that the proposed approaches improve the network performance in terms of latency and consumed energy.

Aerial-Aided Multiaccess Edge Computing: Dynamic and Joint Optimization of Task and Service Placement and Routing in Multilayer Networks

Ubiquity in network coverage is one of the main features of 5G and is expected to be extended to the computing domain in 6G. In order to provide ubiquity in communication and computation, the integration of satellite, aerial and terrestrial networks is foreseen. In particular, the rising amount of applications such as In-Flight Entertainment and Connectivity Services (IFECS) and Software Defined Networking-enabled satellites renders network management more challenging. Moreover, due to the stringent Quality of Service (QoS) requirements, edge computing is vital for these applications. Network performance can be boosted by considering components of the aerial network, like aircrafts, as potential Multi-Access Edge Computing (MEC) nodes. Thus, to cater to the QoS-critical applications, we propose an Aerial-Aided Multi-Access Edge Computing (AA-MEC) architecture that provides a framework for optimal management of computing resources and internet-based services in the sky. Furthermore, we formulate optimization problems to minimize the network latency for the two use cases of providing IFECS to other aircrafts in the sky and providing services for offloading AI/ML-tasks from satellites. Due to the dynamic nature of the satellite and aerial networks, we propose a re-configurable optimization. For the transforming network we continuously identify the optimal MEC node for each application and the optimal path to the destination MEC node. In summary, our results demonstrate, that using AA-MEC improves network latency performance by 10.43% compared to the traditional approach of using only terrestrial MEC nodes, for latency-critical applications such as online gaming. Furthermore, while comparing our proposed dynamic approach with a static one, we record a minimum of 6.7% decrease in flow latency for IFECS and 56.03% decrease for computation offloading.

Resource Management and Reflection Optimization for Intelligent Reflecting Surface Assisted Multi-Access Edge Computing Using Deep Reinforcement Learning

Multi-access edge computing (MEC) enables the computation-intensive and latency-critical application to be processed at the network edge, which reduces the transmission latency and energy consumption. The quality of the wireless channel seriously affects the performance of the edge network. Consequently, the performance of the edge network can be significantly improved from the perspective of communication. The recently advocated intelligent reflecting surface (IRS) intelligently controls the radio propagation environment to improve the quality of wireless communication links. This paper proposes an edge heterogeneous network with the assistance of intelligent reflecting surface. Specifically, the macro base station and small base stations are equipped with MEC servers, and IRS is adopted to provide an additional computation offloading link. The user association, computation offloading and resource allocation, as well as IRS phase shift design are optimized with the aim of minimizing the long-term energy consumption subject to the constraints imposed on quality of service (QoS) and available resources. The challenge of the optimization problem is rooted from the fact that update timescale of user association is different from others. Hence, a two-timescale mechanism is invoked by marrying tools from matching theory and deep reinforcement learning. More specifically, the user association decision takes place in the long timescale. In the short timescale, the computation offloading, resource allocation and IRS phase shift design strategy is performed. The effectiveness of the proposed two-timescale mechanism is verified by the simulation results.

CoreNap: Energy Efficient Core Allocation for Latency-Critical Workloads

In data-center servers, the dynamic core allocation for Latency-Critical (LC) applications can play a crucial role in improving energy efficiency under Service Level Objective (SLO) constraints, allowing cores to enter idle states (i.e., C-states) that consume less power by turning off a part of hardware components of a processor. However, prior studies focus on the core allocation for application threads while not considering cores involved in network packet processing, even though packet processing affects not only response latency but also energy consumption considerably. In this paper, we first investigate the impacts of the explicit core allocation for network packet processing on the tail response latency and energy consumption while running LC applications. We observe that co-adjusting the number of cores for network packet processing along with the number of cores for LC application threads can improve energy efficiency substantially, compared with adjusting the number of cores only for application threads, as prior studies do. In addition, we propose a dynamic core allocation, called <monospace xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">CoreNap</monospace> , which allocates/de-allocates cores for both LC application threads and packet processing. <monospace xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">CoreNap</monospace> measures the CPU-utilization by application threads and packet processing individually, and predicts response latency and power consumption when the combination of core allocation is enforced via a lightweight prediction model. Based on the prediction, <monospace xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">CoreNap</monospace> chooses/enforces the energy-efficient combination of core allocation. Our experimental results show that <monospace xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">CoreNap</monospace> reduces energy consumption by up to 18.6% compared with state-of-the-art study that adjusts cores only for LC application in parallel packet processing environments.

Resource Scheduling in Edge Computing: Architecture, Taxonomy, Open Issues and Future Research Directions

An inflection point in the computing industry is occurring with the implementation of the Internet of Things and 5G communications, which has pushed centralized cloud computing toward edge computing resulting in a paradigm shift in computing. The purpose of edge computing is to provide computing, network control, and storage to the network edge to accommodate computationally intense and latency-critical applications at resource-limited endpoints. Edge computing allows edge devices to offload their overflowing computing tasks to edge servers. This procedure may completely exploit the edge server’s computational and storage capabilities and efficiently execute computing operations. However, transferring all the overflowing computing tasks to an edge server leads to long processing delays and surprisingly high energy consumption for numerous computing tasks. Aside from this, unused edge devices and powerful cloud centers may lead to resource waste. Thus, hiring a collaborative scheduling approach based on task properties, optimization targets, and system status with edge servers, cloud centers, and edge devices is critical for the successful operation of edge computing. This paper briefly summarizes the edge computing architecture for information and task processing. Meanwhile, the collaborative scheduling scenarios are examined. Resource scheduling techniques are then discussed and compared based on four collaboration modes. As part of our survey, we present a thorough overview of the various task offloading schemes proposed by researchers for edge computing. Additionally, according to the literature surveyed, we briefly looked at the fairness and load balancing indicators in scheduling. Finally, edge computing resource scheduling issues, challenges, and future directions have discussed.

Mitigating Trade-Off in Unlicensed Network Optimization Through Machine Learning and Context Awareness

Unlicensed cellular networks are being deployed worldwide by cellular operators to meet the rising data demands. However, the unlicensed band has existing incumbents such as Wi-Fi and radar systems. This creates a highly dynamic environment, making harmonious unlicensed coexistence difficult. Consequently, conventional optimization techniques are not sufficient to offer latency-critical applications and services. A data-driven hybrid optimization approach is necessary for optimal network performance with low convergence times. However, a largely unexplored problem in dense unlicensed network optimization is the accuracy-speed trade-off, that is, achieving high accuracy in optimization objectives with minimal time costs. This work seeks to address this problem through a hybrid optimization approach that combines machine learning and network optimization. It investigates the use of more precise higher-order network feature relationships (NFRs) in optimization formulations and the consequent trade-off that arises between the increase in convergence time (Speed) and the nearness to optimal results (Accuracy). In addition, it demonstrates the relevance of context awareness of network conditions and the traffic environment to mitigate the trade-off. To that end, a context-aware network feature relationship-based optimization (CANEFRO) approach is proposed and validated through decision matrix analysis. The experiments were carried out on a coexistence testbed consisting of both unlicensed LTE standards (LTE-U & LAA) and two Wi-Fi standards (802.11n/ac) on multiple channel bandwidths. In addition, LTE-U & LAA are contrasted on signaling and user data traffic data models and resource block allocation performance. More importantly, CANEFRO demonstrates the impact of the network context on the degree of feature relationship ( <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$2^{nd}$ </tex-math></inline-formula> & <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$3^{rd}$ </tex-math></inline-formula> degree polynomials), objective of optimization (SINR and Capacity), and the network use case (Accuracy vs. Speed). CANEFRO is also used to contrast LTE-U & LAA optimization performance. In particular, the decision matrix analysis demonstrates a higher decision score for LAA by as much as 42% compared to LTE-U.

A Survey on Nongeostationary Satellite Systems: The Communication Perspective

The next phase of satellite technology is being characterized by a new evolution in non-geostationary orbit (NGSO) satellites, which conveys exciting new communication capabilities to provide non-terrestrial connectivity solutions and to support a wide range of digital technologies from various industries. NGSO communication systems are known for a number of key features such as lower propagation delay, smaller size, and lower signal losses in comparison to the conventional geostationary orbit (GSO) satellites, which can potentially enable latency-critical applications to be provided through satellites. NGSO promises a substantial boost in communication speed and energy efficiency, and thus, tackling the main inhibiting factors of commercializing GSO satellites for broader utilization. The promised improvements of NGSO systems have motivated this paper to provide a comprehensive survey of the state-of-the-art NGSO research focusing on the communication prospects, including physical layer and radio access technologies along with the networking aspects and the overall system features and architectures. Beyond this, there are still many NGSO deployment challenges to be addressed to ensure seamless integration not only with GSO systems but also with terrestrial networks. These unprecedented challenges are also discussed in this paper, including coexistence with GSO systems in terms of spectrum access and regulatory issues, satellite constellation and architecture designs, resource management problems, and user equipment requirements. Finally, we outline a set of innovative research directions and new opportunities for future NGSO research.

A Two-Timescale Approach to Mobility Management for Multicell Mobile Edge Computing

Mobile edge computing (MEC) is a promising technology for enhancing the computation capacities and features of mobile users by offloading complex computation tasks to the edge servers. However, mobility poses great challenges on delivering reliable MEC service required for latency-critical applications. First, mobility management has to tackle the dynamics of both user’s location changes and task arrivals that vary in different timescales. Second, user mobility could induce service migration, leading to reliability loss due to the migration delay. In this paper, we propose a two-timescale mobility management framework by joint control of service migration and transmission power to address the above challenges. Specifically, the service migration operates at a large timescale to support user mobility in the multi-cell network, while the power control is performed at a small timescale for real-time task offloading. Their joint control is formulated as an optimization problem aiming at the long-term mobile energy minimization subject to the reliability requirement of computation offloading. To solve the problem, we propose a Lyapunov-based framework to decompose the problem into different timescales, based on which a low-complexity two-timescale online algorithm is developed by exploiting the problem structure. The proposed online algorithm is shown to be asymptotically optimal via theoretical analysis, and is further developed to accommodate the multiuser management. The simulation results demonstrate that our proposed algorithm can significantly improve the energy and reliability performance.

BLOCKCHAIN-ENABLED FOG RESOURCE ACCESS AND GRANTING

Fog computing is a new computing paradigm for meeting ubiquitous massive access and latency-critical applications by moving the processing capability closer to end users. The geographical distribution/floating features with potential autonomy requirements introduce new challenges to the traditional methodology of network access control. In this project, a blockchain-enabled fog resource access and the granting solution is proposed to tackle the unique requirements brought by fog computing. The smart contractual concept is introduced to enable dynamic, and automatic credential generation and delivery for an independent offer of fog resources. Negotiated transaction mechanism supports the fog resource provider to dynamically publish an offer and facilitates the choice of the preferred resource by the end-user. There is no central authority needed to verify your identity, and relieve the processing pressure brought by massive access and singlepoint failure. Our solution can be extended and used in multi-access and especially multi-carrier scenarios in which centralized authorities are absent.

Cloud-Edge Orchestration for Smart Cities: A Review of Kubernetes-based Orchestration Architectures

Edge computing offers computational resources near data-generating devices to enable low-latency access. Especially for smart city contexts, edge computing becomes inevitable for providing real-time services, like air quality monitoring systems. Kubernetes, a popular container orchestration platform, is often used to efficiently manage containerized applications in smart cities. Although it misses essential requirements of edge computing, like network-related metrics for scheduling decisions, it is still considered. This paper analyzes custom cloud-edge architectures implemented with Kubernetes. Specifically, we analyze how essential requirements of edge orchestration in smart cities are solved. Also, shortcomings are identified in these architectures based on the fundamental requirements of edge orchestration. We conduct a literature review to obtain the general requirements of edge computing and edge orchestration for our analysis. We map these requirements to the capabilities of Kubernetes-based cloud-edge architectures to assess their level of achievement. Issues like using network-related metrics and the missing topology-awareness of networks are partially solved. However, requirements like real-time resource utilization, fault-tolerance, and the placement of container registries are in the early stages. We conclude that Kubernetes is an eligible candidate for cloudedge orchestration. When the formerly mentioned issues are solved, Kubernetes can successfully contribute latency-critical, large-scale, and multi-tenant application deployments for smart cities.

Effective Task Scheduling in Critical Fog Applications

Information and technology have witnessed significant improvement with the introduction of Internet of things (IoT) applications, and most of the IoT applications are dependent on the cloud. Cloud computing is assisting IoT applications by providing storage, analysis, and processing services on the cloud. However, Fog computing is the new paradigm that supports the cloud by providing scheduling, resources optimization, and energy optimization services. Scheduling tasks based on MIPs size and prioritizing the tasks with smaller MIPs size first make critical tasks with larger MIPs wait, which ultimately increases the delay and may result in some serious problems. This paper proposes a methodology for critical tasks having large MIPs size by scheduling and prioritizing the tasks based on the nature of the task. The proposed methodology for latency-critical applications reduces latency, energy consumption, and network utilization. This paper proposed a scheduler “Critical task First Scheduler” (CTFS), which schedules tasks depending on the nature of the requests, which are classified as either critical or noncritical. The proposed methodology is implemented in a healthcare scenario, and the simulations are performed in iFogSim simulator. Critical requests, such as emergency notifications, are prioritized and designated as critical, requiring immediate processing. The environment was kept the same for all the approaches that are implemented to demonstrate the effectiveness of the proposed approach. The results of the proposed approach were compared with First Come First Served (FCFS), Shortest Job First (SJF), and cloud-only approaches to demonstrate the effectiveness of the proposed approach in terms of latency, energy consumption, and network utilization. Simulation results show that the proposed CTFS approach outperformed the compared techniques for all three comparison parameters.

Cost and Latency Optimized Edge Computing Platform

Latency-critical applications, e.g., automated and assisted driving services, can now be deployed in fog or edge computing environments, offloading energy-consuming tasks from end devices. Besides the proximity, though, the edge computing platform must provide the necessary operation techniques in order to avoid added delays by all means. In this paper, we propose an integrated edge platform that comprises orchestration methods with such objectives, in terms of handling the deployment of both functions and data. We show how the integration of the function orchestration solution with the adaptive data placement of a distributed key–value store can lead to decreased end-to-end latency even when the mobility of end devices creates a dynamic set of requirements. Along with the necessary monitoring features, the proposed edge platform is capable of serving the nomad users of novel applications with low latency requirements. We showcase this capability in several scenarios, in which we articulate the end-to-end latency performance of our platform by comparing delay measurements with the benchmark of a Redis-based setup lacking the adaptive nature of data orchestration. Our results prove that the stringent delay requisites necessitate the close integration that we present in this paper: functions and data must be orchestrated in sync in order to fully exploit the potential that the proximity of edge resources enables.

Effect of Hyper-Threading in Latency-Critical Multithreaded Cloud Applications and Utilization Analysis of the Major System Resources

Multithreaded latency-critical applications represent an important subset of workloads running on public cloud systems. Most of these systems deploy powerful computing servers including Intel Hyper-Threading processors. Understanding how performance is affected by the consumption of the main system resources is a major concern for cloud providers in order to devise virtualization strategies that improve the system efficiency. With this aim, this paper first characterizes the impact of QPS on tail latency, analyzing different scenarios varying the number of threads and the thread-to-core allocation (single-task and multi-task execution) policy. The characterization study reveals that the performance of some applications does not scale with the number of threads, and the performance of some others is insensitive to the Hyper-Threading technology, so they can be allocated in less physical cores and improve system utilization. Identifying these applications, however, at run-time is challenging. Despite identifying these applications at run-time is challenging, this paper shows that they can be successfully detected at run-time by analyzing the utilization trend of the major system resources. In addition to CPU, we have also studied how assigning the share of each application of other major shared system resources impacts on performance. We outline considerations cloud providers should take into account to improve performance and resource utilization.

Image and Video Coding Techniques for Ultra-low Latency

The next generation of wireless networks fosters the adoption of latency-critical applications such as XR, connected industry, or autonomous driving. This survey gathers implementation aspects of different image and video coding schemes and discusses their tradeoffs. Standardized video coding technologies such as HEVC or VVC provide a high compression ratio, but their enormous complexity sets the scene for alternative approaches like still image, mezzanine, or texture compression in scenarios with tight resource or latency constraints. Regardless of the coding scheme, we found inter-device memory transfers and the lack of sub-frame coding as limitations of current full-system and software-programmable implementations.