Diana Andreea Popescu

Where Has My Time Gone

Time matters. In a networked world, we would like mobile devices to provide a crisp user experience and applications to instantaneously return results. Unfortunately, application performance does not depend solely on processing time, but also on a number of different components that are commonly counted in the overall system latency. Latency is more than just a nuisance to the user, poorly accounted-for, it degrades application performance. In fields such as high frequency trading, as well as in many data centers, latency translates easily to financial losses. Research to date has focused on specific contributions to latency: from improving latency within the network to latency control on the application level. This paper takes an holistic approach to latency, and aims to provide a break-down of end-to-end latency from the application level to the wire. Using a set of crafted experiments, we explore the many contributors to latency. We assert that more attention should be paid to the latency within the host, and show that there is no silver bullet to solve the end-to-end latency challenge in data centers. We believe that a better understanding of the key elements influencing data center latency can trigger a more focused research, improving the user’s quality of experience.

Enabling Fast Hierarchical Heavy Hitter Detection using Programmable Data Planes

Measuring and monitoring network traffic is a fundamental aspect in network management. This poster is a first step towards an SDN solution using an event triggered approach to support advanced monitoring dataplane capabilities. Leveraging P4 programmability, we built a solution to inform a remote controller about the detected hierarchical heavy hitters, thus minimizing control plane overheads.

An Application-Assisted Checkpoint-Restart Mechanism for Java Applications

In this paper we present a novel application-assisted checkpoint-restart mechanism for Java applications. The checkpoint-restart API provides the application developers with full control over what data needs to be check-pointed. The novelty of our system is that it allows different checkpoint periods for different data items. Our implementation makes full use of the Java Reflection API.

Characterizing the impact of network latency on cloud-based applications’ performance

Businesses and individuals run increasing numbers of applications in the cloud. The performance of an application running in the cloud depends on the data center conditions and upon the resources committed to an application. Small network delays may lead to a significant performance degradation, which affects both the user’s cost and the service provider’s resource usage, power consumption and data center efficiency. In this work, we quantify the effect of network latency on several typical cloud workloads, varying in complexity and use cases. Our results show that different applications are affected by fixed and variable latency to differing amounts. These insights into the effect of network latency on different applications have ramifications for workload placement and physical host sharing when trying to reach performance targets.

Multivariate Polynomial Multiplication on GPU

Multivariate polynomial multiplication is a fundamental operation which is used in many scientific domains, for example in the optics code for particle accelerator design at CERN. We present a novel and efficient multivariate polynomial multiplication algorithm for GPUs using floating-point double precision coefficients implemented using the CUDA parallel programming platform. We obtain very good speedups over another multivariate polynomial multiplication library for GPUs (up to 548x), and over the implementation of our algorithm for multi-core machines using OpenMP (up to 7.46x).

Dataset and Reproduction Environment for paper "A First Look At Network Conditions Through The Eyes of PTPmesh"

PTP measurements taken between pairs of VMs in different cloud providers. The dataset is organised as follows: 1. LowMessageFrequency/ contains the data for sections IV A (stats files) and V (messages log files). 2. HighMessageFrequency/ contains the data for section IV B. 3. OtherExperiments/ contains the data for sections III and V. The folders have names according to the conventions used in the paper. The folder environment contains the scripts used to collect the data.

PTPmesh: Data Center Network Latency Measurements Using PTP

Many data center applications are latency-sensitive. Monitoring continuously the network latency and reacting to congestion on a network path is important to ensure that the applications performance does not suffer penalties. We show how to use the Precision Time Protocol (PTP) to infer network latency and packet loss in data centers, and we conduct network latency and packet loss measurements in data centers from different cloud providers, using PTPd, an open-source softwareimplementation of PTP.

Research data and software supporting PTPmesh: Data Center Network Latency Measurements Using PTP

Dataset collected for the publication PTPmesh: Data Center Network Latency Measurements Using PTP