In Mobile Edge Computing (MEC), many tasks require specific service support for execution and in addition, have a dependent order of among the tasks. However, previous works often ignore impact having limited services cached at edge nodes on (dependent) task offloading, thus may lead to an infeasible offloading decision or longer completion time. To bridge gap, this article studies how efficiently offload with (and predetermined) caching. We formally define problem caching (ODT-SC), prove...

In Mobile Edge Computing (MEC), many tasks require specific service support for execution and in addition, have a dependent order of among the tasks. However, previous works often ignore impact having limited services cached at edge nodes on (dependent) task offloading, thus may lead to an infeasible offloading decision or longer completion time. To bridge gap, this paper studies how efficiently offload with (and predetermined) caching. We formally define problem caching (ODT-SC), prove that...

This paper studies reliable teleportation of quantum bits (called qubits) in a data network with multiple sources (S) and destinations (D) as well repeaters. To teleport qubits for SD pair reliably, not only an entanglement path the pair, but also appropriate purification links along is required to ensure that end-to-end (E2E) fidelity established connections high enough.This first work on quantifying E2E fidelity, using this determine critical achieve most resource efficient purification. A...

As the scale of distributed training increases, it brings huge communication overhead in clusters. Some works try to reduce cost through gradient compression or scheduling. However, these methods either downgrade accuracy do not total transmission amount. One promising approach, called in-network aggregation, is proposed mitigate bandwidth bottleneck clusters by aggregating gradients programmable hardware (e.g., Intel Tofino switches). existing solutions mainly implement aggregation fixed...

Traditional networks rely on aggregate routing and decentralized control to achieve scalability. On the contrary, software-defined near optimal network performance policy-based management through per-flow centralized control, which however face scalability challenge due (1) limited TCAM on-die memory for storing forwarding table (2) communication/computation overhead at controller. This paper presents a novel hybrid switching design, integrates traditional SDN purpose of achieving both...

Traditional networks rely on aggregate routing and decentralized control to achieve scalability. On the contrary, software-defined near optimal network performance policy-based management through per-flow centralized control, which, however, face scalability challenge due to: 1) limited ternary content addressable memory on-die for storing forwarding table 2) communication/computation overhead at controller. This paper presents a novel hybrid switching (HS) design, which integrates...

Software-defined networking (SDN) separates the control plane from data to ease network management and provide flexibility in packet routing. The interacts with through an interface that configures forwarding tables, usually including a flow table group table, at each switch. Due high cost power consumption of ternary content addressable memory, commodity switches can only support flow/group tables limited size, which presents serious challenge for SDN scale large networks. One promising...

There are two conventional methods to establish an entanglement connection in a Quantum Data Network (QDN). One is create single-hop links first and then connect them with quantum swapping, the other for-warding one of entangled photons from end via all-optical switching at intermediate nodes directly connection. Since photon easy be lost during long distance transmission, all existing works adopting former method. However, room size network, success probability delivering across multiple...

Network Function Virtualization (NFV) is a new paradigm to enable service innovation through virtualizing traditional network functions. To construct NFV-enabled network, there are two critical requirements: minimizing server deployment cost and satisfying switch resource constraints. However, prior work mostly focuses on the cost, while ignoring constraints (e.g., switch's flow-table size). It thus results in large number of rules switches leads massive control overhead. address this...

Fine-grained flow management is useful in many practical applications, e.g., resource allocation, anomaly detection and traffic engineering. However, it difficult to provide fine-grained for a large number of flows SDNs due switches' limited table capacity. While using wildcard rules can reduce the entries needed, cannot fully ensure all without degrading application performance. In this article, we design implement hybrid rule placement (to be referred as HiFi here after). achieves with...

In a commercial cloud, service providers (e.g., video streaming provider) rent resources from cloud vendors Google Cloud Platform) and provide services to users, making profit the price gap. users acquire by forwarding their requests corresponding servers. practice, as common scenario, traffic dynamics will cause server overload or load-unbalancing. Existing works mainly deal with problem two methods: elastic resource provisioning request updating. Elastic is fast agile solution but may cost...

Quantum Data Networking can significantly transform the landscape of quantum computing by enabling several small computers (QCs) to form a distributed system achieve same power as large computer which is infeasible build. However, this requires state information, in qubits, be exchanged among multiple geographically QCs, and there are many challenges associated with reliably transferring qubits from one QC another efficiently. In paper, we discuss various QDN design options, present main...

Software Defined Network (SDN) is an architectural trend in networking towards the use of centralized controller to get better performance. However, due limited resources (especially flow table size and processing capacity), it may result low-throughput long-delay for a set bursty flows. In this paper, we first combine constraint capacity define Throughput Maximization with Limited Resources (TMLR) problem. Then prove TMLR NP-Hard design approximation algorithm solve The factor proposed also...

The past decades have seen a proliferation of middlebox deployment in various networks, including backbone networks and datacenters. Since network flows to traverse specific service function chains (SFCs) for security performance enhancement, it becomes much complex SFC routing due loops, traffic dynamics scalability requirement. existing solutions may consume many resources (e.g., TCAM) on the data plane lead massive overhead control plane, which decrease networks. Due requirement potential...

To better serve a diversity of flows, load balancing is crucial to ensure operational efficiency. However, previous works for have several disadvantages: 1) limited applicability with sub-flow scheduling (e.g., LetFlow); 2) hash collision ECMP); or 3) transient network congestion due reactive traffic dynamics Hedera and DevoFlow). An important reason the above disadvantages that it difficult provide fully fine-grained flow control in an SDN as table size each switch usually limited. Inspired...

In multi-tenant clouds, requests need to traverse a set of network functions (NFs) in specific order, referred as service function chain (SFC), for security and business logic issues. Due workload dynamics, the central controller cloud needs frequently update SFC routing, so optimize various performance, such load balancing. To achieve effective routing update, we should consider two critical requirements: <italic xmlns:mml="http://www.w3.org/1998/Math/MathML"...

As a simple and low-cost way to obtain enough computing resources, more tenants migrate their tasks the cloud. However, frequent occurrence of abnormal events (e.g., malicious node failures) in cloud will seriously affect tenants' QoS. Conventionally, vendors reduce frequency by deploying auxiliary systems, which requires additional costs increases network complexity. Considering that it is an unrealistic expectation eliminate clouds, this paper proposes complementary scheme alleviate...

Traditional kernel network processing suffers from high delay and overhead, which has become the bottleneck of high-speed networks. A natural method to accelerate packet is bypass stack process packets in user space directly, $e.g$., DPDK. However, due many functions are implemented stack, bypassing means that we need redesign required elsewhere, leading poor compatibility. One promising technology address this problem called eXpress Data Path (XDP), can support high-performance while...

The surging scale of distributed training (DT) incurs significant communication overhead in datacenters, while a promising solution is in-network aggregation (INA). It leverages programmable switches (e.g., Intel Tofino switches) for gradient to accelerate the DT. Due switches' limited on-chip memory size, existing solutions try design sharing mechanism INA. This requires gradients arrive at synchronously, network dynamics make it common asynchronous arrival gradients, resulting being...