The Bridge - Issue 3, 2021 - 22

Feature
Analyzing Service Mesh Performance
VI. SUMMARY
To deploy a service mesh effectively, we need to quantify
application workload characteristics and the utilization
of particular microarchitectures; to assess requirements
for Container Network Interface (CNI) drivers, Open
Virtual Switch (OVS), rules processing, and lookups
between Network Address Translated (NAT) and routed
networks; to consider different layers of service mesh
to be deployed including layer 4 load balancers, ingress
and reverse proxy, number of sidecars and number of
microservices to be supported; to determine hardware
baseline performance for the specific setup; and to set
a quantifiable measure of service mesh deployed with
performance measures mapped to KPIs like throughput
(RPS) and latency.
REFERENCES
[12] Lee, E.J.; Choi, M.H., Moon, D.K., " Enhanced Photovoltaic [1]
Wubin Li, Yves Lemieux, Jing Gao, Zhuofeng Zhao, and Yanbo
Han, " Service Mesh: Challenges, State of the Art, and Future
Research Opportunities, " 2019 IEEE International Conference on
Service-Oriented System Engineering (SOSE), pp. 122-127, 2019.
[2] Janakiram MSV, " Service Mesh - The New Battleground for
The Platform Wars, " Forbes, September 20, 2020. Available:
https://www.forbes.com/sites/janakirammsv/2020/09/20/
service-meshthe-new-battleground-for-the-platformwars/?sh=2412c1243021
[3]
Ektemal Al-Rayis and Heba Kurdi, " Performance Analysis of
Load Balancing Architectures in Cloud Computing, " 20-22 Nov
2013, European Modelling Symposium.
[4] Meshery, August 2021. Available: https://meshery.io
[5] Nighthawk, Getnighthawk, August 2021.
Available: https://getnighthawk.dev
[6] Service Mesh Performance (SMP), August 2021.
Available: https://smp-spec.io
[7] Service Mesh Landscape, Layer5, August 2021.
Available: https://layer5.io/service-mesh-landscape
[8] Calico, Tigera, August 2021.
Available: https://www.tigera.io/project-calico/
[9] Fortio, August 2021. Available: https://fortio.org
[10] Ahmad Yasin, " A Top-Down Method for Performance Analysis
and Counters Architecture, " 2014 International Symposium on
Performance Analysis of Systems and Software (ISPASS),
pp 35-44, 2014.
Lee Calcote is the Founder and CEO
of Layer5, representing the world's
largest collection of service mesh
initiatives. Previously, Calcote stewarded
technology strategy and innovation
across SolarWinds as head of CTO
technology initiatives. He led softwaredefined
data center engineering at
Seagate, delivering predictive analytics
and modern systems management. Calcote held various
leadership positions at Cisco, where he created Cisco's cloud
management platform, pioneered software-defined network
orchestration, and autonomic, remote management services.
Calcote is an author of service mesh books, and a maintainer of
Meshery, Service Mesh Performance, and Service Mesh Interface
open source projects. He serves on Cisco's technical advisory
board and as chair of the Cloud Native Computing Foundation's
TAG for Networking. (Layer5, lee.calcote@layer5.io)
Mrittika Ganguli Mrittika Ganguli is a
Principal Engineer and Director, Cloud
Native Pathfinding in Intel's NEX Group.
Her area of expertise is in hardware and
software platform management, network
and storage processing control and data
plane, cloud orchestration, telemetry
QOS and scheduling Architecture.
Service Mesh and Cloud Native
performance benchmarking and optimization is her area of focus
now. (Intel, mrittika.ganguli@intel.com)
Sunku Ranganath is a solutions
architect for Edge Compute at Intel
with over 11 years of experience. His
work for the last few years has been
enabling service meshes for Telecom
domain, designing, building, integrating,
benchmarking NFV based reference
architectures using Kubernetes &
OpenStack components with focus on
enabling hardware features, closed loop automation, telemetry.
Sunku is very active in open-source projects, being a maintainer
for CNCF Service Mesh Performance & CollectD projects, served
as Technical Steering Committee member of OPNFV (now
Anuket) & IEEE initiatives such as Future Networks Edge Service
Platform & IEEE P1916-1 standard being a IEEE senior member.
(Intel, sunkuranganath@ieee.org)
Otto Van Der Schaaf is a principal
software engineer at Red Hat. His focus
is on proxy software with a special
interest in measuring and optimizing
performance. He is a maintainer of
the CNCF envoyproxy/nighthawk and
Service Mesh Performance projects,
as well as a committer and project
management committee member for
the Apache PageSpeed (incubating) and Traffic Server projects.
(Red Hat, ovanders@redhat.com)
THE BRIDGE

https://www.forbes.com/sites/janakirammsv/2020/09/20/service-meshthe-new-battleground-for-the-platformwars/?sh=2412c1243021 https://www.forbes.com/sites/janakirammsv/2020/09/20/service-meshthe-new-battleground-for-the-platformwars/?sh=2412c1243021 https://www.forbes.com/sites/janakirammsv/2020/09/20/service-meshthe-new-battleground-for-the-platformwars/?sh=2412c1243021 https://meshery.io/ https://getnighthawk.dev/ https://smp-spec.io/ https://www.layer5.io/service-mesh-landscape https://www.tigera.io/project-calico/ https://www.fortio.org https://hkn.ieee.org/

The Bridge - Issue 3, 2021

Table of Contents for the Digital Edition of The Bridge - Issue 3, 2021