Posco Tso

Dr. Posco Tso


Research Focus & Metrics

Academic Career Length

19Years (2008-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
5 July 2021 in IEEE Transactions on Network and Service Management
Verified
Journal Article
Closed access
XZXiaoquan Zhang
Xiaoquan Zhang
LCLin Cui
Lin Cui
PTPosco Tso
Posco Tso
WJWei Jia
Wei Jia
Since heavy flows account for a significant fraction of network traffic, being able to predict heavy flows has benefited many network management applications for mitigating link congestion, scheduling of network capacity, exposing network attacks and so on. Existing machine learning based predictors are largely implemented on the control plane of Software Defined Networking (SDN) paradigm. As a result, frequent communication between the control and data planes can cause unnecessary overhead and additional delay in decision making. In this paper, we present pHeavy , a machine learning based scheme for predicting heavy flows directly on the programmable data plane, thus eliminating network overhead and latency to SDN controller. Considering the scarce memory and limited computation capability in the programmable data plane, pHeavy includes a packet processing pipeline which deploys pre-trained decision tree models for in-network prediction. We have implemented pHeavy in both bmv2 software switch and P4 hardware switch (i.e., Barefoot Tofino). Evaluation results demonstrate that pHeavy has achieved 85% and 98% accuracy after receiving the first 5 and 20 packets of a flow respectively, while being able to reduce the size of decision tree by 5.4x on average. More importantly, pHeavy can predict heavy flows at line rate on the P4 hardware switch.

6 June 2016 in Electronics
Verified
Journal Article
Open access
PDF file
WHWajdi Hajji
Wajdi Hajji
PTPosco Tso
Posco Tso
Nowadays, Internet-of-Things (IoT) devices generate data at high speed and large volume. Often the data require real-time processing to support high system responsiveness which can be supported by localised Cloud and/or Fog computing paradigms. However, there are considerably large deployments of IoT such as sensor networks in remote areas where Internet connectivity is sparse, challenging the localised Cloud and/or Fog computing paradigms. With the advent of the Raspberry Pi, a credit card-sized single board computer, there is a great opportunity to construct low-cost, low-power portable cloud to support real-time data processing next to IoT deployments. In this paper, we extend our previous work on constructing Raspberry Pi Cloud to study its feasibility for real-time big data analytics under realistic application-level workload in both native and virtualised environments. We have extensively tested the performance of a single node Raspberry Pi 2 Model B with httperf and a cluster of 12 nodes with Apache Spark and HDFS (Hadoop Distributed File System). Our results have demonstrated that our portable cloud is useful for supporting real-time big data analytics. On the other hand, our results have also unveiled that overhead for CPU-bound workload in virtualised environment is surprisingly high, at 67.2%. We have found that, for big data applications, the virtualisation overhead is fractional for small jobs but becomes more significant for large jobs, up to 28.6%.

8 May 2023
Verified
Conference Paper
Open access
CCChen Chen
Chen Chen
LNLars Nagel
Lars Nagel
LCLin Cui
Lin Cui
PTPosco Tso
Posco Tso
Serverless edge computing uses an event-driven model in which Internet-of-Things (IoT) services are run in short-lived, stateless containers only when invoked, leading to significant reduction of resource utilization. However, a cold-start of a container can take up to several seconds which significantly degrades the response time of serverless applications. Container caching can mitigate the cold-start problem at the cost of extra computing resources which violates the spirit of serverless computing. Therefore, we need to balance the cold-start overheads with the extra resource utilization for serverless edge computing. Nevertheless, the diverse ranges of containers lead to different cold-start overheads, resource consumption and invocation frequencies and these characteristics of containers are largely overlooked by existing caching policies. In this paper, we study the request distribution and caching problem for serverless edge computing. We devise an online request distribution algorithm with performance guarantee and present an adaptive caching policy which incorporates container frequency, container size and cold-start time. Via real-system implementation, the superiority of the proposed algorithm is verified by comparing with existing caching policies, including fixed caching and histogram based policies. Our results show that the proposed algorithm reduces both the average response time and cold-start frequency by a factor of 3 compared to current approaches.

21 January 2022 in Journal of Systems Architecture
Verified
Journal Article
Closed access
WLWenjie Liang
Wenjie Liang
LCLin Cui
Lin Cui
PTPosco Tso
Posco Tso

31 August 2016 in IEEE Transactions on Parallel and Distributed Systems
Verified
Journal Article
Open access
LCLin Cui
Lin Cui
PTPosco Tso
Posco Tso
Dimitrios PezarosDimitrios Pezaros
Dimitrios Pezaros
WJWei Jia
Wei Jia
WZWei Zhao
Wei Zhao
Policies play an important role in network configuration and therefore in offering secure and high performance services especially over multi-tenant Cloud Data Center (DC) environments. At the same time, elastic resource provisioning through virtualization often disregards policy requirements, assuming that the policy implementation is handled by the underlying network infrastructure. This can result in policy violations, performance degradation and security vulnerabilities. In this paper, we define PLAN, a PoLicy-Aware and Network-aware VM management scheme to jointly consider DC communication cost reduction through Virtual Machine (VM) migration while meeting network policy requirements. We show that the problem is NP-hard and derive an efficient approximate algorithm to reduce communication cost while adhering to policy constraints. Through extensive evaluation, we show that PLAN can reduce topology-wide communication cost by 38 percent over diverse aggregate traffic and configuration policies.

23 October 2023 in IEEE Transactions on Services Computing
Verified
Journal Article
Closed access
MQMimi Qian
Mimi Qian
LCLin Cui
Lin Cui
PTPosco Tso
Posco Tso
YDYuhui Deng
Yuhui Deng
WJWei Jia
Wei Jia
Network measurement is essential for efficient network management and operations. In-band network telemetry (INT) offers fine-grained per-device per-packet information which could provide full-visibility for networks. However, the existing solutions fall short in achieving high accuracy, generality, and low overhead simultaneously. To address this limitation, we introduce OffsetINT to meet these three criteria. The key idea of OffsetINT is to use minimal bits to carry collected states during monitoring, which is based on our observation that the value of telemetry states are usually very close (e.g., the time of adjacent arrival packets) or small (e.g., only a few tens of microseconds for processing latency) for most of the time in real networks. Instead of embedding complete values of state in packets, OffsetINT optimizes bit usage by encoding an offset (using fewer bits), which is carried in-band by passing packets to the end-hosts for recovery and analysis. We theoretically derive the bounds of bandwidth mitigation for OffsetINT . We have implemented OffsetINT in both P4 hardware switches (with Intel Tofino ASIC) and BMv2. Expensive evaluation results show that OffsetINT can achieve an accuracy of up to 100% compared to the original INT while reducing INT bandwidth by up to 48%.