Artem M. Chirkin

Artem M. Chirkin

Auto-generated profile, based on publicly available information

Artem M. Chirkin

NVIDIA, Zürich, Switzerland

Research Focus & Metrics

Academic Career Length

14Years (2013-Present)

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

Representative Publications

View All Publications
1 February 2018 in Cities
Verified
Journal Article
Open access
JMJohannes Mueller
Johannes Mueller
HLHangxin Lu
Hangxin Lu
ACArtem Chirkin
Artem Chirkin
BKBernhard Klein
Bernhard Klein
GSGerhard Schmitt
Gerhard Schmitt

1 October 2017 in Future Generation Computer Systems
Verified
Journal Article
Closed access
ACArtem M. Chirkin
Artem M. Chirkin
ABAdam S.Z. Belloum
Adam S.Z. Belloum
Sergey KovalchukSergey Kovalchuk
Sergey Kovalchuk
MMMarc X. Makkes
Marc X. Makkes
MMMikhail A. Melnik
Mikhail A. Melnik
DNDenis A. Nasonov
Denis A. Nasonov

1 January 2013 in Procedia Computer Science
Verified
Journal Article
Open access
VKV.V. Krzhizhanovskaya
V.V. Krzhizhanovskaya
NMN.B. Melnikova
N.B. Melnikova
ACA.M. Chirkin
A.M. Chirkin
SIS.V. Ivanov
S.V. Ivanov
ABAlexander V. Boukhanovsky
Alexander V. Boukhanovsky
PSPeter Sloot
Peter Sloot
We present a decision support system for flood early warning and disaster management. It includes the models for data- driven meteorological predictions, for simulation of atmospheric pressure, wind, long sea waves and seiches; a module for optimization of flood barrier gates operation; models for stability assessment of levees and embankments, for simulation of city inundation dynamics and citizens evacuation scenarios. The novelty of this paper is a coupled distributed simulation of surface and subsurface flows that can predict inundation of low-lying inland zones far away from the edge of the flooded area, as observed in St. Petersburg city during the floods. All the models are wrapped as software services in the CLAVIRE platform for urgent computing, which provides workflow management and resource orchestration

1 November 2014
Verified
Conference Paper
Closed access
ACArtem M. Chirkin
Artem M. Chirkin
ABA. S. Z. Belloum
A. S. Z. Belloum
Sergey KovalchukSergey Kovalchuk
Sergey Kovalchuk
MMMarc X. Makkes
Marc X. Makkes
Estimation of the execution time is an important part of the workflow scheduling problem. The aim of this paper is to highlight common problems in estimating the workflow execution time and propose a solution that takes into account the complexity and the randomness of the workflow components and their runtime. The solution proposed in this paper addresses the problems at different levels from task to workflow, including the error measurement and the theory behind the estimation algorithm. The proposed estimation algorithm can be integrated easily into a wide class of schedulers as a separate module. We use a dual stochastic representation, characteristic / distribution functions, in order to combine tasks' estimates into the overall workflow makespan. Additionally, we propose the workflow reductions - the operations on a workflow graph that do not decrease the accuracy of the estimates, but simplify the graph structure, hence increasing the performance of the algorithm.

7 May 2016
Verified
Conference Paper
Open access
ACArtem M. Chirkin
Artem M. Chirkin
RKReinhard König
Reinhard König
This work presents a concept of interactive machine learning in a human design process. An urban design problem is viewed as a multiple-criteria optimization problem. The outlined feature of an urban design problem is the dependence of a design goal on a context of the problem. We model the design goal as a randomized fitness measure that depends on the context. In terms of multiple-criteria decision analysis (MCDA), the defined measure corresponds to a subjective expected utility of a user. In the first stage of the proposed approach we let the algorithm explore a design space using clustering techniques. The second stage is an interactive design loop; the user makes a proposal, then the program optimizes it, gets the user's feedback and returns back the control over the application interface.

22 September 2025 in Proceedings of the ACM on Management of Data
Verified
Journal Article
Open access
JXJingyi Xi
Jingyi Xi
CMChenghao Mo
Chenghao Mo
BKBen Karsin
Ben Karsin
ACArtem Chirkin
Artem Chirkin
MLMingqin Li
Mingqin Li
MZMinjia Zhang
Minjia Zhang
Vector search and database systems have become a keystone component in many AI applications. While many prior research has investigated how to accelerate the performance of generic vector search, emerging AI applications require running more sophisticated vector queries efficiently, such as vector search with attribute filters. Unfortunately, recent filtered-ANNS solutions are primarily designed for CPUs, with few exploration and limited performance of filtered-ANNS that take advantage of the massive parallelism offered by GPUs. In this paper, we present VecFlow, a novel high-performance vector filtered search system that achieves unprecedented high throughput and recall while obtaining low latency for filtered-ANNS on GPUs. We propose a novel label-centric indexing and search algorithm that significantly improves the selectivity of ANNS with filters. In addition to algorithmic level optimization, we provide architecture-aware optimizations for VecFlow's functional modules, effectively supporting both small batch and large batch queries, and single-label and multi-label query processing. Experimental results on NVIDIA A100 GPU over several public available datasets validate that VecFlow achieves 5 million QPS for recall 90%, outperforming state-of-the-art CPU-based solutions such as Filtered-DiskANN by up to 135 times. Alternatively, VecFlow can easily extend its support to high recall 99% regime, whereas strong GPU-based baselines plateau at around 80% recall.