A5012901455- Software-Defined Networks and 5G
- Network Traffic and Congestion Control
- Caching and Content Delivery
- Cloud Computing and Resource Management
- Advanced Optical Network Technologies
- IoT and Edge/Fog Computing
- Peer-to-Peer Network Technologies
- Interconnection Networks and Systems
- Software System Performance and Reliability
- Distributed and Parallel Computing Systems
- Anomaly Detection Techniques and Applications
- Network Packet Processing and Optimization
- Advanced MIMO Systems Optimization
- Optimization and Search Problems
- Telecommunications and Broadcasting Technologies
- Text and Document Classification Technologies
- Computational Geometry and Mesh Generation
- Advanced Neural Network Applications
- Advanced Queuing Theory Analysis
- Reservoir Engineering and Simulation Methods
- Network Security and Intrusion Detection
- Robotics and Automated Systems
- Cooperative Communication and Network Coding
- Distributed systems and fault tolerance
- Formal Methods in Verification
IBM Research - Haifa
2015-2024
IBM (United States)
2010
Technion – Israel Institute of Technology
1996-2006
IBM Research - Tokyo
2002
We consider the problem of routing connections with quality service (QoS) requirements across networks when information available for making decisions is inaccurate. Such uncertainty about actual state a network component arises naturally in number different environments. The goal route selection process then to identify path that most likely satisfy QoS requirements. For end-to-end delay guarantees, this intractable. However, we show by decomposing constraint into local constraints,...
The placement of virtual machines (VMs) on a cluster hosts under multiple constraints, including administrative (security, regulations) resource-oriented (capacity, energy), and QoS-oriented (performance) is highly complex task. We define new high-availability property for VM; when VM marked as k-resilient, long there are up to k host failures, it should be guaranteed that can relocated non-failed without relocating other VMs. Together with Hardware Predictive Failure Analysis live...
Media use cases for emergency services require mission-critical levels of reliability the delivery media-rich services, such as video streaming. With upcoming deployment fifth generation (5G) networks, a wide variety applications and with heterogeneous performance requirements are expected to be supported, any migration 5G networks presents significant challenges in quality service (QoS), operators. This paper novel SliceNet framework, based on advanced customizable network slicing address...
Network slicing has emerged as a major new networking paradigm for meeting the diverse requirements of various vertical businesses in virtualised and softwarised 5G networks. SliceNet is project EU Infrastructure Public Private Partnership (5G PPP) focuses on network cornerstone technology networks, addresses associated challenges managing, controlling orchestrating services users especially sectors, thereby maximising potential infrastructures their by leveraging advanced software cognitive...
We investigate the problem of optimal resource allocation for end-to-end QoS requirements on unicast paths and multicast trees. Specifically, we consider a framework in which is based local at each network link, associated with link cost function that increases severity requirement. Accordingly, address how to partition an requirement into requirements, such overall minimized. establish efficient (polynomial) solutions both connections. These results provide required foundations...
This article considers the problem of routing connections with QoS requirements across networks, when information available for making decisions is inaccurate. uncertainty about actual state a network component arises naturally in number different environments, which are reviewed paper. The goal route selection process then to identify path that most likely satisfy requirements. For end delay guarantees, this intractable. However we show by decomposing end-to-end constraint into local...
In this paper we study problems related to supporting unicast and multicast connections with quality of service (QoS) requirements. We investigate the problem optimal routing resource allocation in context performance-dependent costs. each network element can offer several QoS guarantees, associated a different cost. This is natural extension commonly used hi-criteria model, where link single delay framework simple yet strong enough model many practical interesting networking problems. An...
User-transparent live migration is one of the most interesting features Virtual Machine (VM) environments. Current live-migration technologies require that VM retains its IP network address; therefore, are typically restricted to movement within an subnet. The growing number portable computing devices has led development mobility solutions enable uninterrupted connectivity while moving between different subnets. In this paper we study application current approaches cross-subnet...
With the growing reliance on ubiquitous availability of IT systems and services, these become more global, scaled, complex to operate. To maintain business viability, service providers must put in place reliable cost efficient operations support. Artificial Intelligence for Operations (AIOps) is a promising technology alleviating operational complexity services. AIOps platforms utilize big data, machine learning other advanced analytics technologies enhance with proactive actionable dynamic...
Mobile edge computing offers a myriad of opportunities to innovate and introduce novel applications, thereby enhancing user experiences considerably.A critical issue extensively investigated in this domain is efficient deployment Service Function Chains (SFCs) across the physical network, spanning from cloud.This problem known be NPhard.As result its practical importance, there significant interest development high-quality sub-optimal solutions.In paper, we consider propose near-optimal...
Consistent hashing is a central building block in many networking applications, such as maintaining connection affinity of TCP flows. However, current consistent solutions do not ensure full consistency under arbitrary changes or scale poorly terms memory footprint, update time and key lookup complexity. We present AnchorHash, scalable fully-consistent algorithm. AnchorHash achieves high rate, low footprint time. formally establish its strong theoretical guarantees, an advanced...
In this paper, we study problems related to supporting unicast and multicast connections with quality of service (QoS) requirements. We investigate the problem optimal routing resource allocation in context performance dependent costs. context, each network element can offer several QoS guarantees, associated a different cost. This is natural extension commonly used bi-criteria model, where link single delay framework simple yet strong enough model many practical interesting networking...
We investigate the problem of optimal resource allocation for end to-end QoS requirements on unicast paths and multicast trees. Specifically, we consider a framework in which is based local at each network link, associated with link cost function that increases severity requirement. Accordingly, address how to partition an end-to-end requirement into requirements, such overall minimized. establish efficient (polynomial) solutions both connections. These results provide required foundations...
Cloud Computing in general and Virtualized Infrastructure Provisioning particular, are significant trends with the potential to increase agility lower costs of IT. An emerging cloud service is a virtual server shop, that allows customers order appliances be delivered virtually on cloud. Like physical shops, want customize ordered products, e.g., have them pre-installed their desired applications pre-configured. Global providers need create customized virtual-server disk images deliver time...
Fifth Generation (5G) networks are envisioned to be fully autonomous in accordance the ETSI-defined Zero touch network and Service Management (ZSM) concept. To this end, purpose-specific Machine Learning (ML) models can used manage control physical as well virtual resources a way that is compliant slice Level Agreements (SLAs), while also boosting revenue of underlying operator(s). This because specially designed trained ML both proactive very effective against management issues induce...
Given a large variety of resources and billing contracts offered by today's cloud providers, customers face nontrivial optimization challenge for their application workloads. A number works are dealing with either selection or resource types selection. We argue that the largest cost savings to elastic workloads result from jointly optimizing heterogeneous To this end, we introduce novel control management framework formulate problem called Heterogeneous Resource Reservation (HRR). evaluate...
The problems related to supporting multicast connections with quality of service (QoS) requirements are studied. We investigate the problem optimal resource allocation in context performance dependent costs. In this each network element can offer several QoS guarantees, associated a different cost. This is natural extension commonly used bi-criteria model, where link single delay and framework simple yet strong enough model many practical interesting networking problems. fundamental under...
Hybrid switching combines a high-bandwidth optical circuit switch in parallel with low-bandwidth electronic packet switch. It presents an appealing solution for scaling datacenter architectures. Unfortunately, it does not fit many traffic patterns produced by typical applications, and particular the skewed that involve highly intensive one-to-many many-to-one communications.
Mobile edge computing offers ultra-low latency, high bandwidth, and reliability. Thus, it can support a plethora of emerging services that be placed in close proximity to the user. One fundamental problems this context is maximizing benefit from placement networked services, while meeting bandwidth latency constraints. In study, we propose an adaptive predictive resource allocation strategy for virtual-network function comprising at mobile edge. Our study focuses on service provider's under...