| 1. |
- Abelev, Betty, et al.
(författare)
-
Measurement of prompt J/psi and beauty hadron production cross sections at mid-rapidity in pp collisions at root s=7 TeV
- 2012
-
Ingår i: Journal of High Energy Physics. - 1029-8479. ; :11
-
Tidskriftsartikel (refereegranskat)abstract
- The ALICE experiment at the LHC has studied J/psi production at mid-rapidity in pp collisions at root s = 7 TeV through its electron pair decay on a data sample corresponding to an integrated luminosity L-int = 5.6 nb(-1). The fraction of J/psi from the decay of long-lived beauty hadrons was determined for J/psi candidates with transverse momentum p(t) > 1,3 GeV/c and rapidity vertical bar y vertical bar < 0.9. The cross section for prompt J/psi mesons, i.e. directly produced J/psi and prompt decays of heavier charmonium states such as the psi(2S) and chi(c) resonances, is sigma(prompt J/psi) (p(t) > 1.3 GeV/c, vertical bar y vertical bar < 0.9) = 8.3 +/- 0.8(stat.) +/- 1.1 (syst.)(-1.4)(+1.5) (syst. pol.) mu b. The cross section for the production of b-hadrons decaying to J/psi with p(t) > 1.3 GeV/c and vertical bar y vertical bar < 0.9 is a sigma(J/psi <- hB) (p(t) > 1.3 GeV/c, vertical bar y vertical bar < 0.9) = 1.46 +/- 0.38 (stat.)(-0.32)(+0.26) (syst.) mu b. The results are compared to QCD model predictions. The shape of the p(t) and y distributions of b-quarks predicted by perturbative QCD model calculations are used to extrapolate the measured cross section to derive the b (b) over bar pair total cross section and d sigma/dy at mid-rapidity.
|
|
| 2. |
- Abelev, Betty, et al.
(författare)
-
Underlying Event measurements in pp collisions at root s=0.9 and 7 TeV with the ALICE experiment at the LHC
- 2012
-
Ingår i: Journal of High Energy Physics. - 1029-8479. ; :7
-
Tidskriftsartikel (refereegranskat)abstract
- We present measurements of Underlying Event observables in pp collisions at root s = 0 : 9 and 7 TeV. The analysis is performed as a function of the highest charged-particle transverse momentum p(T),L-T in the event. Different regions are defined with respect to the azimuthal direction of the leading (highest transverse momentum) track: Toward, Transverse and Away. The Toward and Away regions collect the fragmentation products of the hardest partonic interaction. The Transverse region is expected to be most sensitive to the Underlying Event activity. The study is performed with charged particles above three different p(T) thresholds: 0.15, 0.5 and 1.0 GeV/c. In the Transverse region we observe an increase in the multiplicity of a factor 2-3 between the lower and higher collision energies, depending on the track p(T) threshold considered. Data are compared to PYTHIA 6.4, PYTHIA 8.1 and PHOJET. On average, all models considered underestimate the multiplicity and summed p(T) in the Transverse region by about 10-30%.
|
|
| 3. |
- Abelev, Betty, et al.
(författare)
-
Long-range angular correlations on the near and away side in p-Pb collisions at root S-NN=5.02 TeV
- 2013
-
Ingår i: Physics Letters. Section B: Nuclear, Elementary Particle and High-Energy Physics. - : Elsevier BV. - 0370-2693. ; 719:1-3, s. 29-41
-
Tidskriftsartikel (refereegranskat)abstract
- Angular correlations between charged trigger and associated particles are measured by the ALICE detector in p-Pb collisions at a nucleon-nucleon centre-of-mass energy of 5.02 TeV for transverse momentum ranges within 0.5 < P-T,P-assoc < P-T,P-trig < 4 GeV/c. The correlations are measured over two units of pseudorapidity and full azimuthal angle in different intervals of event multiplicity, and expressed as associated yield per trigger particle. Two long-range ridge-like structures, one on the near side and one on the away side, are observed when the per-trigger yield obtained in low-multiplicity events is subtracted from the one in high-multiplicity events. The excess on the near-side is qualitatively similar to that recently reported by the CMS Collaboration, while the excess on the away-side is reported for the first time. The two-ridge structure projected onto azimuthal angle is quantified with the second and third Fourier coefficients as well as by near-side and away-side yields and widths. The yields on the near side and on the away side are equal within the uncertainties for all studied event multiplicity and p(T) bins, and the widths show no significant evolution with event multiplicity or p(T). These findings suggest that the near-side ridge is accompanied by an essentially identical away-side ridge. (c) 2013 CERN. Published by Elsevier B.V. All rights reserved.
|
|
| 4. |
- Bogdanov, Kirill, 1987-
(författare)
-
Enabling Fast and Accurate Run-Time Decisions in Geo-Distributed Systems : Better Achieving Service Level Objectives
- 2018
-
Doktorsavhandling (övrigt vetenskapligt/konstnärligt)abstract
- Computing services are highly integrated into modern society and used by millions of people daily. To meet these high demands, many popular services are implemented and deployed as geo-distributed applications on top of third-party virtualized cloud providers. However, the nature of such a deployment leads to variable performance. To deliver high quality of service, these systems strive to adapt to ever-changing conditions by monitoring changes in state and making informed run-time decisions, such as choosing server peering, replica placement, and redirection of requests. In this dissertation, we seek to improve the quality of run-time decisions made by geo-distributed systems. We attempt to achieve this through: (1) a better understanding of the underlying deployment conditions, (2) systematic and thorough testing of the decision logic implemented in these systems, and (3) by providing a clear view of the network and system states allowing services to make better-informed decisions. First, we validate an application’s decision logic used in popular storage systems by examining replica selection algorithms. We do this by introducing GeoPerf, a tool that uses symbolic execution and modeling to perform systematic testing of replica selection algorithms. GeoPerf was used to test two popular storage systems and found one bug in each. Then, using measurements across EC2, we observed persistent correlation between network paths and network latency. Based on these observations, we introduce EdgeVar, a tool that decouples routing and congestion based changes in network latency. This additional information improves estimation of latency, as well as increases the stability of network path selection. Next, we introduce Tectonic, a tool that tracks an application’s requests and responses both at the user and kernel levels. In combination with EdgeVar, it decouples end-to-end request completion time into three components of network routing, network congestion, and service time. Finally, we demonstrate how this decoupling of request completion time components can be leveraged in practice by developing Kurma, a fast and accurate load balancer for geo-distributed storage systems. At runtime, Kurma integrates network latency and service time distributions to accurately estimate the rate of Service Level Objective (SLO) violations, for requests redirected between geo-distributed datacenters. Using real-world data, we demonstrate Kurma’s ability to effectively share load among datacenters while reducing SLO violations by a factor of up to 3 in high load settings or reducing the cost of running the service by up to 17%. The techniques described in this dissertation are important for current and future geo-distributed services that strive to provide the best quality of service to customers while minimizing the cost of operating the service.
|
|
| 5. |
- Bogdanov, Kirill, et al.
(författare)
-
Fast and accurate load balancing for geo-distributed storage systems
- 2018
-
Ingår i: SoCC 2018 - Proceedings of the 2018 ACM Symposium on Cloud Computing. - New York, NY, USA : Association for Computing Machinery (ACM). - 9781450360111 ; , s. 386-400
-
Konferensbidrag (refereegranskat)abstract
- The increasing density of globally distributed datacenters reduces the network latency between neighboring datacenters and allows replicated services deployed across neighboring locations to share workload when necessary, without violating strict Service Level Objectives (SLOs). We present Kurma, a practical implementation of a fast and accurate load balancer for geo-distributed storage systems. At run-time, Kurma integrates network latency and service time distributions to accurately estimate the rate of SLO violations for requests redirected across geo-distributed datacenters. Using these estimates, Kurma solves a decentralized rate-based performance model enabling fast load balancing (in the order of seconds) while taming global SLO violations. We integrate Kurma with Cassandra, a popular storage system. Using real-world traces along with a geo-distributed deployment across Amazon EC2, we demonstrate Kurma’s ability to effectively share load among datacenters while reducing SLO violations by up to a factor of 3 in high load settings or reducing the cost of running the service by up to 17%.
|
|
| 6. |
|
|
| 7. |
|
|
| 8. |
- Bogdanov, Kirill, 1987-
(författare)
-
Reducing Long Tail Latencies in Geo-Distributed Systems
- 2016
-
Licentiatavhandling (övrigt vetenskapligt/konstnärligt)abstract
- Computing services are highly integrated into modern society. Millions of people rely on these services daily for communication, coordination, trading, and accessing to information. To meet high demands, many popular services are implemented and deployed as geo-distributed applications on top of third party virtualized cloud providers. However, the nature of such deployment provides variable performance characteristics. To deliver high quality of service, such systems strive to adapt to ever-changing conditions by monitoring changes in state and making run-time decisions, such as choosing server peering, replica placement, and quorum selection.In this thesis, we seek to improve the quality of run-time decisions made by geo-distributed systems. We attempt to achieve this through: (1) a better understanding of the underlying deployment conditions, (2) systematic and thorough testing of the decision logic implemented in these systems, and (3) by providing a clear view into the network and system states which allows these services to perform better-informed decisions.We performed a long-term cross datacenter latency measurement of the Amazon EC2 cloud provider. We used this data to quantify the variability of network conditions and demonstrated its impact on the performance of the systems deployed on top of this cloud provider.Next, we validate an application’s decision logic used in popular storage systems by examining replica selection algorithms. We introduce GeoPerf, a tool that uses symbolic execution and lightweight modeling to perform systematic testing of replica selection algorithms. We applied GeoPerf to test two popular storage systems and we found one bug in each.Then, using traceroute and one-way delay measurements across EC2, we demonstrated persistent correlation between network paths and network latency. We introduce EdgeVar, a tool that decouples routing and congestion based changes in network latency. By providing this additional information, we improved the quality of latency estimation, as well as increased the stability of network path selection.Finally, we introduce Tectonic, a tool that tracks an application’s requests and responses both at the user and kernel levels. In combination with EdgeVar, it provides a complete view of the delays associated with each processing stage of a request and response. Using Tectonic, we analyzed the impact of sharing CPUs in a virtualized environment and can infer the hypervisor’s scheduling policies. We argue for the importance of knowing these policies and propose to use them in applications’ decision making process.
|
|
| 9. |
- Bogdanov, Kirill, et al.
(författare)
-
The Nearest Replica Can Be Farther Than You Think
- 2015
-
Ingår i: Proceedings of the ACM Symposium on Cloud Computing 2015. - New York, NY, USA : Association for Computing Machinery (ACM). ; , s. 16-29
-
Konferensbidrag (refereegranskat)abstract
- Modern distributed systems are geo-distributed for reasons of increased performance, reliability, and survivability. At the heart of many such systems, e.g., the widely used Cassandra and MongoDB data stores, is an algorithm for choosing a closest set of replicas to service a client request. Suboptimal replica choices due to dynamically changing network conditions result in reduced performance as a result of increased response latency. We present GeoPerf, a tool that tries to automate the process of systematically testing the performance of replica selection algorithms for geodistributed storage systems. Our key idea is to combine symbolic execution and lightweight modeling to generate a set of inputs that can expose weaknesses in replica selection. As part of our evaluation, we analyzed network round trip times between geographically distributed Amazon EC2 regions, and showed a significant number of daily changes in nearestK replica orders. We tested Cassandra and MongoDB using our tool, and found bugs in each of these systems. Finally, we use our collected Amazon EC2 latency traces to quantify the time lost due to these bugs. For example due to the bug in Cassandra, the median wasted time for 10% of all requests is above 50 ms.
|
|
| 10. |
- Bogdanov, Kirill, et al.
(författare)
-
Toward Automated Testing of Geo-Distributed Replica Selection Algorithms
- 2015
-
Ingår i: Proceedings of the 2015 ACM Conference on Special Interest Group on Data Communication. - New York, NY, USA : Association for Computing Machinery (ACM). ; , s. 89-90
-
Konferensbidrag (refereegranskat)abstract
- Many geo-distributed systems rely on a replica selection algorithms to communicate with the closest set of replicas. Unfortunately, the bursty nature of the Internet traffic and ever changing network conditions present a problem in identifying the best choices of replicas. Suboptimal replica choices result in increased response latency and reduced system performance. In this work we present GeoPerf, a tool that tries to automate testing of geo-distributed replica selection algorithms. We used GeoPerf to test Cassandra and MongoDB, two popular data stores, and found bugs in each of these systems.
|
|
