Zevenet EE 4.x. Performance Benchmark.

Similar documents
Apple. Massive Scale Deployment / Connectivity. This is not a contribution

Scaling Internet TV Content Delivery ALEX GUTARIN DIRECTOR OF ENGINEERING, NETFLIX

Alpha Anywhere Standard Application Server

Load Balancing Nginx Web Servers with OWASP Top 10 WAF in Azure

Deployment Guide AX Series with Oracle E-Business Suite 12

Load Balancing Nginx Web Servers with OWASP Top 10 WAF in AWS

Results of a stress test between DAS servers.

Datacenter Wide- area Enterprise

A10 Thunder ADC with Oracle E-Business Suite 12.2 DEPLOYMENT GUIDE

NetScaler 2048-bit SSL Performance

Load Balancing Microsoft IIS. Deployment Guide v Copyright Loadbalancer.org, Inc

DEPLOYMENT GUIDE A10 THUNDER ADC FOR EPIC SYSTEMS

Load Balancing Microsoft IIS. Deployment Guide v Copyright Loadbalancer.org

DevCentral Basics: Application Delivery Services PRESENTED BY:

How to Make the Client IP Address Available to the Back-end Server

SE Memory Consumption

Benchmark Study: A Performance Comparison Between RHEL 5 and RHEL 6 on System z

STORAGE MADE EASY ENTERPRISE FILE FABRIC

Datacenter Wide- area Enterprise

Citrix NetScaler Traffic Management

Load Balancing Web Servers with OWASP Top 10 WAF in Azure

SE Memory Consumption

STATEFUL TCP/UDP traffic generation and analysis

TCP Tuning for the Web

Diffusion TM 5.0 Performance Benchmarks

EMQ-2.0 Benchmark Report

SEDA: An Architecture for Well-Conditioned, Scalable Internet Services

BlackBerry AtHoc Networked Crisis Communication Capacity Planning Guidelines. AtHoc SMS Codes

Load Balancing Web Servers with OWASP Top 10 WAF in AWS

Scaling Web Service. Capacity planning. CS144: Web Applications

Tuning NGINX for high performance. Nick Shadrin

A Rational software Whitepaper 05/25/03. IBM Rational Rapid Developer Scalability and Performance Benchmark

WINDOWS SERVER 2012 ECHOSTREAMS FLACHESAN2

FortiTester Handbook VERSION 2.5.0

Meet the Increased Demands on Your Infrastructure with Dell and Intel. ServerWatchTM Executive Brief

Tuning NGINX for high performance. Nick Shadrin

IBM Daeja ViewONE Virtual Performance and Scalability

Performance and Scalability Benchmark: Siebel CRM Release 7 on HP-UX Servers and Oracle9i Database. An Oracle White Paper Released October 2003

Ingo Brenckmann Jochen Kirsten Storage Technology Strategists SAS EMEA Copyright 2003, SAS Institute Inc. All rights reserved.

Performance Test Report. Prepared by Kentico Software in April 2010

Estimate performance and capacity requirements for Access Services

Oracle Cloud Metric Reference for Oracle Infrastructure Monitoring

Intel Enterprise Edition Lustre (IEEL-2.3) [DNE-1 enabled] on Dell MD Storage

A Library and Proxy for SPDY

Network Design Considerations for Grid Computing

Análise e Modelagem de Desempenho de Sistemas de Computação: Component Level Performance Models of Computer Systems

Почему IBM POWER8 оптимальная платформа для PostgreSQL

Reliable Distributed Messaging with HornetQ

Citrix NetScaler Administration Training

F5 Big-IP LTM Configuration: HTTPS / WSS Offloading

Load Balancing Microsoft Terminal Services. Deployment Guide v Copyright Loadbalancer.org, Inc

PT APPLICATION FIREWALL BASED ON PT UNIFIED CHASSIS

Distil Networks & HAProxy Integration Guide

Emulex LPe16000B 16Gb Fibre Channel HBA Evaluation

Cisco Wide Area Application Services (WAAS) Mobile

SCALE AND SECURE MOBILE / IOT MQTT TRAFFIC

Distil Networks & F5 Networks Integration Guide

Speeding up Linux TCP/IP with a Fast Packet I/O Framework

NetScaler for Apps and Desktops CNS-222; 5 Days; Instructor-led

Xytech MediaPulse Equipment Guidelines (Version 8 and Sky)

ERDAS APOLLO PERFORMANCE BENCHMARK ECW DELIVERY PERFORMANCE OF ERDAS APOLLO VERSUS ESRI ARCGIS FOR SERVER

Consolidating Microsoft SQL Server databases on PowerEdge R930 server

Ch. 7: Benchmarks and Performance Tests

User manual. CMS performance SINGLE SERVER SCENARIOS. Preliminary version: English

Crescando: Predictable Performance for Unpredictable Workloads

Improving the DragonFlyBSD Network Stack

Benefits of full TCP/IP offload in the NFS

DNS Survival Guide. Artyom Gavrichenkov

Finding the Needle in the Haystack

Slicing a Network. Software-Defined Network (SDN) FlowVisor. Advanced! Computer Networks. Centralized Network Control (NC)

Xytech MediaPulse Equipment Guidelines (Version 8 and Sky)

Leverage the Citrix WANScaler Software Client to Increase Application Performance for Mobile Users

SYSTEM REQUIREMENTS M.APP ENTERPRISE

Elastic Load Balancing

W H I T E P A P E R. Comparison of Storage Protocol Performance in VMware vsphere 4

Scaling Up Performance Benchmarking

Oracle IaaS, a modern felhő infrastruktúra

TRex Realistic Traffic Generator

FUJITSU Software Interstage Information Integrator V11

VMware AirWatch Content Gateway Guide for Windows

Load Balancing VMware Horizon View. Deployment Guide v Copyright Loadbalancer.org, Inc

AMP in the Enterprise Open Source Confidence. March 2005

Load Balancing Microsoft OCS Deployment Guide v Copyright Loadbalancer.org

Load Balancing Bloxx Web Filter. Deployment Guide v Copyright Loadbalancer.org

Load Balancing Microsoft Remote Desktop Services. Deployment Guide v Copyright Loadbalancer.org, Inc

System Requirements Overview

SharePoint 2010 Technical Case Study: Microsoft SharePoint Server 2010 Social Environment

WHITE PAPER: BEST PRACTICES. Sizing and Scalability Recommendations for Symantec Endpoint Protection. Symantec Enterprise Security Solutions Group

Performance Testing: A Comparative Study and Analysis of Web Service Testing Tools

Distributing Web-based Content Management System - FERweb

TALK THUNDER SOFTWARE FOR BARE METAL HIGH-PERFORMANCE SOFTWARE FOR THE MODERN DATA CENTER WITH A10 DATASHEET YOUR CHOICE OF HARDWARE

Daniel A. Menascé, Ph. D. Dept. of Computer Science George Mason University

How to Configure Virus Scanning in the Firewall for FTP Traffic

AMD: WebBench Virtualization Performance Study

Performance Analysis of iscsi Middleware Optimized for Encryption Processing in a Long-Latency Environment

Load Balancing OKI DICOM-Embedded Printers. Deployment Guide v Copyright Loadbalancer.org

Load Balancing RSA Authentication Manager. Deployment Guide v Copyright Loadbalancer.org, Inc

Understanding of basic networking concepts (routing, switching, VLAN, firewall functionality)

vcloud Automation Center Reference Architecture vcloud Automation Center 5.2

Microsoft Exchange 2000 Front-End Server and SMTP Gateway Hardware Scalability Guide. White Paper

Transcription:

Zevenet EE 4.x Performance Benchmark www.zevenet.com

Performance Benchmark Zevenet EE 4.x January, 2017 Content Table 1. Benchmark Scenario 2. Benchmark Cases 2.1. L4xNAT Profile 2.2. HTTP Profile with HTTP Listener 2.3. HTTP Profile with HTTPS Offload Listener 3. Results Summary / 1

1. Benchmark Scenario The scenario is created with 2 connected networks using the ZNA hardware Appliance with the latest Zevenet Enterprise Edition 4.x package optimized for ZNA EE 3300. The networking environment is: 1. Service Network: 172.16.1.0/24 a. Eth1: 172.16.1.1 in ZNA Virtual Services b. Clients: 172.16.1.0/24 2. Backends Network: 172.16.2.0/24 a. Eth2: 172.16.2.1 in ZNA b. Backends: 172.16.2.0/24 The ZNA appliance provides a Zevenet EE 4.1.7 version with a CPU Intel Core i5 660 (2 cores with HT) 3.33G and 4 GB of RAM DDR3. / 2

The client and backends host limits haven't been reached during the benchmarking tests. The client stress application used was 'wrk' which is a HTTP benchmarking tool that generates heavy HTTP and HTTPS requests against the listening service of the load balancer. The backends were configured with a simple nginx web service that received the client requests and responded an empty web page (0 bytes of html body and 233 bytes in HTTP headers). 2. Benchmark Cases 2.1. L4xNAT Profile Farm Configuration in Zevenet Farm Profile Configured parameters L4xNAT Farm Virtual IP: 172.16.1.1 Virtual Port: 80 Protocol: TCP NAT Type: snat Command executed in the client root@client:/#./wrk -d30 -t10 -c360 http://172.16.1.1/ The execution time was 30 secs long and simulates 10 threads with a concurrency of 360 each, the result is shown below: Running 30s test @ http://172.16.1.1/ 10 threads and 360 connections Thread Stats Avg Stdev Max +/- Stdev Latency 2.98ms 9.13ms 226.61ms 95.51% Req/Sec 26.92k 6.40k 58.56k 70.65% 7986968 requests in 30.10s, 1.73GB read Requests/sec: 265387.32 Transfer/sec: 58.96MB / 3

CPU usage in ZNA While the test was running, the CPU in the ZNA was used as follows: As it's shown above, all CPUs reached an average of 28% and were moving more than 265,000 request per seconds, HTTP requests are managed smoothly. 2.2. HTTP Profile with HTTP Listener Farm Configuration in Zevenet Farm Profile HTTP Configured parameters Farm Virtual IP: 172.16.1.1 Virtual Port: 80 Number of working threads: 2,000 Farm Listener: HTTP Command executed in the client root@client:/#./wrk -d30 -t10 -c360 http://172.16.1.1/ The execution time was 30 secs long and simulates 10 threads with a concurrency of 360 each, the result is shown below: / 4

Running 30s test @ http://172.16.1.1 10 threads and 360 connections Thread Stats Avg Stdev Max +/- Stdev Latency 26.96ms 26.58ms 681.01ms 91.49% Req/Sec 1.45k 538.42 10.41k 73.15% 431308 requests in 30.10s, 95.82MB read Requests/sec: 14329.43 Transfer/sec: 3.18MB CPU usage in ZNA While the test is running, the CPU in the ZNA is used as follows: As it's show, all CPUs of ZNA reach the 100% of usage but the HTTP requests are managed smoothly. 2.3. HTTP Profile with HTTPS Offload Listener Farm Configuration in Zevenet Farm Profile HTTP Configured parameters Farm Virtual IP: 172.16.1.1 Virtual Port: 443 Number of working threads: 2,000 Farm Listener: HTTPS Ciphers: High Security / 5

Command executed in the client root@client:/#./wrk -d30 -t10 -c360 https://172.16.1.1/ The execution time is 30 secs and simulates 10 threads with a concurrency of 360, the output result is shown below: Running 30s test @ https://172.16.1.1/ 10 threads and 360 connections Thread Stats Avg Stdev Max +/- Stdev Latency 31.25ms 26.22ms 457.78ms 61.57% Req/Sec 1.25k 374.61 3.39k 76.81% 371846 requests in 30.02s, 98.57MB read Requests/sec: 12388.38 Transfer/sec: 3.28MB CPU usage in ZNA While the test is running, the CPU in the ZNA is used as follows: As it's show, all CPUs of ZNA reach the 100% of usage but the SSL requests are managed smoothly. / 6

3. Results Summary The following shows the performance benchmark results for all cases over the ZNA hardware appliance and web traffic: Farm Profile Requests in 30s Requests per second Avg Latency Estimated HTTP users handled per second** L4 xnat* 7,986,968 265,387 2.98ms 10,615,480 HTTP 431,308 14,329 26.96ms 143,290 HTTPS 371,846 12,388 31.25ms 123,880 * CPU was never saturated (~25%) ** 1 RPS is able to simulate 10 concurrent users To sum up, in the table above we can see how the ZNA load balancer is able to manage more than 265k request per second with a L4xNAT profile using just the 25% of the CPU performance. With such performance, the L4xNAT is able to handle more than 10.6 millions of web users per second. The performance reached with a HTTP farm profile is able to handle more than 143 thousands of web users per second and near 124 thousands of web users per second with high security ciphers in HTTPS layer. / 7