Related papers: SLO beyond the Hardware Isolation Limits
Hardware accelerators, such as those based on GPUs and FPGAs, offer an excellent opportunity to efficiently parallelize functionalities. Recently, modern embedded platforms started being equipped with such accelerators, resulting in a…
In this paper, we propose a solution to proactively mitigate Distributed Denial-of-Service attacks in 5G core network slicing using slice isolation. Network slicing is one of the key technologies that allow 5G networks to offer dedicated…
We propose a novel computing runtime that exposes remote compute devices via the cross-vendor open heterogeneous computing standard OpenCL and can execute compute tasks on the MEC cluster side across multiple servers in a scalable manner.…
Split learning (SL) enables collaborative training of large language models (LLMs) between resource-constrained edge devices and compute-rich servers by partitioning model computation across the network boundary. However, existing SL…
We consider robust resource allocation of services in Clouds. More specifically, we consider the case of a large public or private Cloud platform that runs a relatively small set of large and independent services. These services are…
Future Exascale systems will feature massive parallelism, many-core processors and heterogeneous architectures. In this scenario, it is increasingly difficult for HPC applications to fully and efficiently utilize the resources in system…
GPUs are vastly underutilized, even when running resource-intensive AI applications, as GPU kernels within each job have diverse resource profiles that may saturate some parts of a device while often leaving other parts idle. Colocating…
Today we see the use of the Internet of Things (IoT) in various application domains such as healthcare, smart homes, smart cars, and smart-x applications in smart cities. The number of applications based on IoT and cloud computing is…
Large language models (LLMs) are becoming increasingly capable at small parameter scales. At the same time, conventional cloud-centric deployment introduces challenges around data privacy, latency, and cost that are acute in operational…
The variety of existing cloud services creates a challenge for service providers to enforce reasonable Software Level Agreements (SLA) stating the Quality of Service (QoS) and penalties in case QoS is not achieved. To avoid such penalties…
A pivotal attribute of 5G networks is their capability to cater to diverse application requirements. This is achieved by creating logically isolated virtual networks, or slices, with distinct service level agreements (SLAs) tailored to…
In hosting environments such as IaaS clouds, desirable application performance is usually guaranteed through the use of Service Level Agreements (SLAs), which specify minimal fractions of resource capacities that must be allocated for use…
Consider a wireless network where each communication link has a minimum bandwidth quality-of-service requirement. Certain pairs of wireless links interfere with each other due to being in the same vicinity, and this interference is modeled…
Modern user-facing latency-sensitive web services include numerous distributed, intercommunicating microservices that promise to simplify software development and operation. However, multiplexing of compute resources across microservices is…
Common implementations of core memory allocation components, like the Linux buddy system, handle concurrent allocation/release requests by synchronizing threads via spin-locks. This approach is clearly not prone to scale with large thread…
Network slicing has emerged as a key concept in 5G systems, allowing Mobile Network Operators (MNOs) to build isolated logical networks (slices) on top of shared infrastructure networks managed by Infrastructure Providers (InP). Network…
Consumer robotics demands consolidation of safety-critical control, perception pipelines, and user applications on shared multicore platforms. While static partitioning hypervisors provide hardware-enforced isolation, directly transplanting…
Memory is a critical design consideration in current data-intensive DNN accelerators, as it profoundly determines energy consumption, bandwidth requirements, and area costs. As DNN structures become more complex, a larger on-chip memory…
The networking community continues to create new technologies and update existing ones to improve the quality, reliability, and "tailorability" of data networks. However, whenever Internet service providers attempt to productize…
Systolic arrays and shared-L1-memory manycore clusters are commonly used architectural paradigms that offer different trade-offs to accelerate parallel workloads. While the first excel with regular dataflow at the cost of rigid…