Presentation on theme: "BTH’s Research in NV, NFV and Cloud Networking"— Presentation transcript:
1BTH’s Research in NV, NFV and Cloud Networking GENI Nordic Meeting, Stockholm,Kurt TutschkuWith Patrik Arlos, Anders Carlsson, Dragos Ilie and Markus FiedlerBlekinge Institute of Technology (BTH), Faculty of Computing Department of Communication Systems (DIKO)
2Capacities at BTH Blekinge Institute of Technology 7200 students; 500 staffemphasis on applied information technology and inno- vation for sustainable growth in industry and societystrong industrial environment in communication industry, both legacy (Ericsson, Telenor Sverige) and startup (CompuVerde, HyperIsland, CityNetwork, ….)network: 1GBit; upgrade to full 10GBit (in 2015)
3Capacities at BTH’s DIKO Department Department of Communication Systems (DIKO)focus on future network/FI architectures and technologies, Quality of Experience, Cloud computing, performance evaluation, wireless communications, Internet of Things and securitycurrently four professors, four senior lecturers, two university adjuncts and 10 Ph.D. studentsCurrent and past involvements in Future Internet projectsCurrent FI projects: XIFI (eXperimental Infrastructures for the Fu-ture Internet, EU), Queen (EU, Celtic plus), ETSI’s Industry Specifi-cation Group (ISG) for Network Function Virtualization (NFV), FI-PPP FI-STAR (EU), ENGENSEC (EU), BTH’s CloudLabSelected past contributions to FI projects: Akari (J), G-Lab (Ger), Mevico (Celtic, EU), PlanetLabEurope, Future Internet Assembly, FI-PPP setup (AT representative)
4BTH CloudLab Started in early 2014 Integration effort for BTH’s FI, NV, NFV, Cloud and SDN researchIntegrated projects and labs: XIFI, DIKO’s Network Performance Lab, ENGENSECHardware:XIFI: 4x Dell PowerEdge 715 (AMD, 128 cores, 512GB Ram, 5TByte disk)ENGENSEC: 48 cores (8boxes; Intel I7); future AMD Opteron128 coresNPL: e.g. Endace DAG 4.3GE x4, DAG 4.2GE x2, DAG 3.5 x4, DAG 3.6 x4Software:OpenStack (XIFI; ENGENSEC: Havana)
5XIFI (eXperimental Infrastructures for the Future Internet, EU)
6Possible use of GEs between IF-PPP Cloud environment and UE BTH’s XIFI TestbedBTH’s XIFI TestbedXIFI adapterDPMINTASFront-end monitoring in NPL:Back-end moni- toring in Cloud LabBTH’s XIFI-enhanced CloudLab running Generic Enablers (GEs)UE executing FI-PPP applicationsMPMonitoring on network layerMonitoring on user layer and client controlPossible use of GEs between IF-PPP Cloud environment and UELink to SUNET/GÉANT network
7Educating the Next gener-ation experts in Cyber Security (ENGENSEC) Objective: create new Master’s program in IT Security as response on current and emerging cybersecurity threats by educating next generation expertsFunding organization: EU Tempus programNumber or participants: 21Participating countries: Sweden (coordinator), Poland, Latvia, Greece, Germany, Ukraine, RussiaProject activities:Defining framework of joint Master’s program, Cloud-based security lab development, Development of the joint course curriculum, Develop new and further develop existing courses, Teacher training, Effective quality control ensured and project management, Dissemination of new Master’s programs benefits, Give pilot courses in a summer school, Prepare for participating Universities to launching new Master’s program
8Direct Involvement of BTH in FI-PPP FI-STAR = one out of five Call-2 FI-PPP use casesBTH’s roleMajor Swedish participant (with significant labs)Requirement engineering (co-chair of FI-STAR WP 1)Validation (Co-chair FI-STAR WP6)Functional testingQuality of Service (QoS) measurementsQuality of Experience (QoE) assessmentHealth Technology AssessmentBTH’s work is strongly focused on Generic Enablers (GE) and their performanceSynergy with XIFI: Hosting would provide full control and unique QoS measurement facilities
9A Very Brief View on Network Function Virtualization (NFV) Kurt TutschkuBlekinge Institute of Technology (BTH), Faculty of ComputingDepartment of Communication Systems (DIKO)
10What is Network Function Virtualization (NFV)? Aims at network operators!Transform network architecture and operation by applying standard IT virtualization technologyMembers: >250 companies; only few academics (5); member since Jan. 2013Amongst other: work on future curricular
11Move this box into the cloud! Example: BRAS – Broadband Residential Access ServerMove this box into the cloud!
12Suggested PoC by SK Telecom Example: Service Chaining in NFV for Video AccelerationSuggested PoC by SK Telecom
14Rules of Thumb, Educated Guesses or Scientific Results? Initial Evaluation: Virtualization Concepts and Their Rough PerformanceRules of Thumb, Educated Guesses or Scientific Results?
15A Metric for Isolation and Trans-parency of Virtual Elements Kurt TutschkuBlekinge Institute of Technology (BTH), Faculty of ComputingDepartment of Communication Systems (DIKO)With acknowledgements to the definitions and descriptions of M. Fiedler (BTH) and D. Stezenbach (University of Vienna)
16Scope and Causes of Reduced Virtualization Features? Sharing ResourcesServer (Host Machine)CPUMemoryI/OVirtual Machine MonitorVirtualMachineGuest OSVirtual applianceVirtual applianceVirtual I/OVirtual MemoryVirtual CPUMain cause for reduced quality of virtualization is resource sharing!(Typically) “atomic” resources: only a single request can be served at a time.However, requests might arrive in parallel (from other VEs due to sharing)Concurrency is resolved by serialization. But, this might introduce additional delay (jitter) for the deferred request.Thought experiment: two virtual appliances, arbitrary schedulingBut, severity depends on “tolerable” delay and in particular on the delay variance.Does this happen in real life?
17Experiment: Sharing among Virtual Routers Set-up:Server: consumer hardware (Intel Core 2 Duo E8500, 4GB RAM, Ubuntu 12.10); network interfaces: 2x1Gbit/s 100Mbit/sVirtual router appliances: Ubuntu 12.10, XEN or VirtualBox ; packet forwarding using vSwitch; four appliances usedMeasurement traffic: 4 parallel UDP streams; 120B Frame size (Ethernet); CBR traffic: inter-packet time 61µs (per stream),15,65Mbit/s per Flow, 62,62Mbit/s totalBe aware: these are data packets but in general this can be extended to signaling/control request
18Experiment: Comparison of ingress and egress Packet sequence:Average throughputIngress (all flows)Egress (all flows) Observations:Ingress: strict round-robinEgress: arbitrary packet orderIngress (all flows)Egress (all flows)Throughput variation is indicator for reduced isolation and transparency!Methodology: comparison of ingress with egress (independent of traffic type)Implementation: compare coefficient of variation at ingress and egress
19Power of the Metric: Comparison Virtualization Technologies – Use of VirtualBox instead of Xen VirtualBox introduces less variation than Xen(our current assumption: this is due to VirtualBox not using the complex vSwitch)Attention: metric does not analyze why a specific virtualization technology has a better isolation/transparency!Focus of metric is on enabling a comparison!VirtualBoxXen