This document discusses high performance Java communication solutions for financial services applications. It introduces Java Fast Sockets and FastMPJ, which provide extremely low-latency shared memory and remote direct memory access communication respectively. These solutions enable direct performance increases, cost reduction, and productivity gains for applications such as high frequency trading, market data distribution, and high performance cloud computing. The document provides an overview of the proposed solutions and includes performance benchmark results demonstrating microsecond-level latencies.

1. Solutions for Financial Services
High Speed Java Communications
PRODUCTS
Java Fast Sockets. High perfor- FastMPJ. High Performance
BENEFITS mance Java socket implementation Message Passing Java implemen-
• Extreme low-latency that supersedes standard Java tation. High performance access to
• Record throughput sockets. Features: both local and remote memory
• Direct integration  Low latency shared memory (LDMA & RDMA). Features:
• The fastest Java TCP/IP communication  High performance intra-node
alternative  High speed networks support communication
• Cost reduction and (e.g., InfiniBand)  Efficient RDMA transfers over
productivity increase InfiniBand and RoCE
 User-transparent for all Java
applications  Ideal for multicore servers,
APPLICATIONS cluster and cloud computing
 Direct performance increase
• High Frequency Trading
 Efficient replacement of  Scalability up to thousands of
• Market Data
TCP/IP stack cores
Distribution
 Fast zero-copy protocol  Highly productive development
• High Performance
Computing and maintenance
• Big Data Processing  Fully portable, as Java
• High Performance Cloud
Computing
PERFORMANCE RESULTS
Testbed: Sandy Bridge E5-2643 3.30 GHz quad-core with 1600DDR3 memory,
PCI Gen3, InfiniBand FDR ConnectX-3 NIC.

2. OVERVIEW OF THE PROPOSED SOLUTION COMMUNICATIONS WITH LOW JITTER
Application  JFS on shared memory:
Sockets
“Fast”
TCP/IP Protocols
High Performance Network Driver
High Performance Network
 Direct Java Fast Sockets integration Mean = 95% =
5% = 0.262 0.283 0.305
 User/application transparent (without Min = 0.236
Median 90% = 99% =
source code modification) since it imple- = 0.28 0.299 0.315
ments the standard JVM socket API.
 FastMPJ based Development
 Development of specific communications
module
 Fault-tolerant and portable
Metrics in microseconds
 Straightforward and high level API,
extremely productive development.  FastMPJ over InfiniBand:
 FastMPJ code example:
import mpi.*;
public class Echo{
public static void main(String args[]) throws Exception {
args=MPI.Init(args);
int myId = MPI.COMM_WORLD.Rank();
int size = MPI.COMM_WORLD.Size();
int peer = size - 1 - myId; Mean = 1.204 95% = 2.577
5% = 0.959
CustomContainer data = new CustomContainer(); 90% = 2.225 99% = 2.828
if(myId==0){
data.set(myData);
MPI.COMM_WORLD.Send(data,MPI.Object,peer,messageId);
System.out.println(“Sending "+data+" to Process "+peer);
} else {
MPI.COMM_WORLD.Recv(data,MPI.Object,peer,messageId);
System.out.println(“Receiving "+data+" from Process"+peer);
}
MPI.Finalize(); Metrics in microseconds
}
}
CONTACT
 Why Message-Passing? Email: info@javafastcomms.com
 Client-Server: Service oriented. Limited WWW: http://www.javafastcomms.com/
scalability
http://www.fastmpj.com/
 Message queueing: Data oriented. Scala-
http://www.javafastsockets.com/
ble but limited performance.
 Message passing: Process oriented. High
performance and scalability. USPTO Patent protected technology developed at the:
SERVICES PROVIDED
 Turnkey solutions
 Full support and development of FastMPJ &
JFS-based solutions With full support by:
 Low latency Java consultancy and training