SlideShare a Scribd company logo
TCP Offload Engine- SoftCore IP


                             INT-1010
                   TCP Offload Engine

Product brief, features and benefits summary

Highly customizable hardware IP block. Easily portable to
ASIC flow, Xilinx or Altera FPGAs

INT-1010 is highly flexible that is customizable for layer-3, layer-
4, layer-5 network infrastructure and network security systems
applications. It is recommended for use in, among others, high
performance Servers, NICs, SAN/NAS and data center applications. It
provides the key IP building block for a single high performance Giga
bit Ethernet ASIC/ASSP/FPGA.

INT-1010 provides capability for enterprises to differentiate their
Network security and Network infrastructure appliances from others

INT-1010 can process TCP/IP, for in-line network security appliances,
sessions in both directions, simultaneously, at full G-bit rate. This
relieves the host CPU from costly TCP/IP buffer stack execution and
maintenance tasks.

INT-1010 can process TCP/IP sessions and has the capability to
process other relevant protocols such as FTP, UDP, ICMP, TFTP etc on
all inbound as well as outbound traffic simultaneously without
compromising performance at sustained Gigabit speeds.


  • Ideal for high performance                      • Scalable MAC Rx FIFOs and Tx
 and mid performance                               FIFOs make it ideal for
 specialized, differentiable                       optimizing system performance.
 ASICs or FPGAs for Network
 security or Network                                • Hardware implementation of
 infrastructure applications                       TCP/IP stacks’ control plane
                                                   and data plane.
  • Less than 4000 Xilinx
 slices, Altera ALMs or 150,000                     • Hardware implementation of
 ASIC gates + on-chip memory                       ARP protocol.

  • Fully integrated 100 M                         ARP table creation, deletion
 bit/1-G bit high performance                      management (optional)
 EMAC.

                    Intelop Corporation www.intelop.com
                    4800 Great America Pkwy. Ste-231
                    Santa Clara, CA. 95054.    Ph: 408-496-0333, Fax: 408-496-0444
TCP Offload Engine- SoftCore IP


 • Customizable for IP-protocol                   live TCP sessions.
only.                                             Customizable, depending upon
                                                  on-chip memory availability.
 • On-chip DDR or SSRAM memory
controller which can address                       • Extendable to 256K TCP
from 4K Bytes to 4 MB Bytes on                    sessions. Internal Memory
chip or 256 MB off chip                           dependent.
memories (User customizable,
optional)                                          • Wire-speed 2Gbps
                                                  performance at full duplex
 • Many trade-offs for some
functions performed in                             • Can process up to 256K
hardware or software                              connections per second

 • Configurable Packet buffers,                    • TCP + IP check sum
session table buffers On-chip                     generation and check performed
or Off-chip memories, attached                    in hardware in less than 6
DDR I/II interface. Depending                     clks (30 ns at 200 MHz) vs 1-2
on system, performance,                           us by typical software TCP-
ASIC/FPGA size requirements->                     stack
User Customizable, (optional)
                                                   • Connection set up and tear
 • Interfaces directly to GMII,                   down/termination
RGMII, MII external 10/100/
1000 Mbit Phy interfaces                           • User programmable Session
                                                  table parameters
 • Architecture can be scaled
up to 10-G bits                                    • Dedicated set of hardware
                                                  Timers for each TCP/IP session
 • Customizable to handle jumbo                   or customizable for sharing
frames                                            stale sessions.

 • Generic bus interface for                       • Multiple ‘slot storage’ for
Local Processor control. ARM,                     fragmented packets. More slots
PPC, MIPS and other CPU                           allocated when more On-chip
interfaces available                              Memory available. Self-
                                                  checking available memory
 • User programmable/                             logic. (optional)
prioritize-able interrupts
                                                   • Out of sequence packet
 • Performs connection/session                    detection/storage and
management                                        Reassembly/Segmentation
                                                  (optional)
 • Monitors, Stores, Maintains
and processes more than 4024                       • RDMA- Direct Data placement
                                                  in Applications buffer at full
                  Intelop Corporation www.intelop.com
                  4800 Great America Pkwy. Ste-231
                  Santa Clara, CA. 95054.    Ph: 408-496-0333, Fax: 408-496-0444
TCP Offload Engine- SoftCore IP

wire speed without CPU->                           • Accommodates future
reduces CPU’s buffer copy time                    Specifications changes.
and utilization by 95%
                                                   • Basic mini API available for
 • Future Proof- Flexible                         PPC architecture. Others
implementation of TCP Offload                     OSs/CPUs also available.




                        Simplified Block Diagram




                  Intelop Corporation www.intelop.com
                  4800 Great America Pkwy. Ste-231
                  Santa Clara, CA. 95054.    Ph: 408-496-0333, Fax: 408-496-0444
TCP Offload Engine- SoftCore IP


Specifications brief:
  • Functionality Proven in multiple IDS/IPS appliances
  • Complete header and flag processing of TCP/IP
    sessions in hardware    accelerates by 10 x – 50 x
  • TCP Offload Engine- 2G b/s Wire-speed performance
  • Scalable to 10 G b/s
  • TCP + IP check sum- hardware
  • TCP segmentation/reassembly in hardware
  • Multiple ‘slot storage’ for fragmented packets
  • Out of sequence packet detection/storage/Reassembly
  • Accelerate security processing, Storage Networking-
    TCP
  • RDMA- Data placement in Applications buffer
-> reduces CPU utilization by 90 %
  • Future Proof- Flexible implementation of TCP Offload
  • Accommodates future Specifications changes.
  • TOE MS-chimney architecture compatible
  • Customizable. Source code; Verilog

  • Detailed specs available under NDA




                        Intelop Corporation www.intelop.com
                        4800 Great America Pkwy. Ste-231
                        Santa Clara, CA. 95054.    Ph: 408-496-0333, Fax: 408-496-0444
TCP Offload Engine- SoftCore IP


Xilinx                                                   Altera




              iTOE                         - FPGA
                           Altera Stratix III,
                           IV, V

                                                 Stratix - 2




                        /ALMs
                        ~2256      ~3256         ~1256    16- 256




                         Alter-vqm,


                                           Verilog


                                            Verilog


    /Altera
                               Quartus 8.1




                                  Quartus 8.1



               Intelop Corporation www.intelop.com
               4800 Great America Pkwy. Ste-231
               Santa Clara, CA. 95054.    Ph: 408-496-0333, Fax: 408-496-0444

More Related Content

What's hot (20)

PDF
Unleashing End-to_end TLS Security Leveraging NGINX with Intel(r) QuickAssist...
Michelle Holley
 
PDF
Xilinx vs Intel (Altera) FPGA performance comparison
Roy Messinger
 
PDF
BKK16-303 96Boards - TV Platform
Linaro
 
PDF
Introduction to nfv movilforum
videos
 
PDF
Intel(r) Quick Assist Technology Overview
Michelle Holley
 
PPT
Upgrade Your Broadcast System to PCIe Gen2
Altera Corporation
 
PPTX
Overview of Nios II Embedded Processor
Altera Corporation
 
PDF
1 intro to_dpdk_and_hw
videos
 
PDF
5 pipeline arch_rationale
videos
 
PDF
6 profiling tools
videos
 
PPT
Sparc t4 systems customer presentation
solarisyougood
 
PPTX
Revisit DCA, PCIe TPH and DDIO
Hisaki Ohara
 
PPTX
The FlexTiles Development Platform offers Dual FPGA for 3D SoC Prototyping
FlexTiles Team
 
PDF
Hpe Proliant DL20 Gen10 Server Datasheet
美兰 曾
 
PPTX
SNAPDRAGON SoC Family and ARM Architecture
Abdullaziz Tagawy
 
PDF
IGLOO2 Microsemi FPGA
Samira Riki
 
PDF
Rhino labs Prese4th ntation At FPGA Camp, Santa Clara, CA
FPGA Central
 
PDF
Session 8,9 PCI Express
Subhash Iyer
 
PDF
ODSA Use Case - SmartNIC
ODSA Workgroup
 
PDF
What are latest new features that DPDK brings into 2018?
Michelle Holley
 
Unleashing End-to_end TLS Security Leveraging NGINX with Intel(r) QuickAssist...
Michelle Holley
 
Xilinx vs Intel (Altera) FPGA performance comparison
Roy Messinger
 
BKK16-303 96Boards - TV Platform
Linaro
 
Introduction to nfv movilforum
videos
 
Intel(r) Quick Assist Technology Overview
Michelle Holley
 
Upgrade Your Broadcast System to PCIe Gen2
Altera Corporation
 
Overview of Nios II Embedded Processor
Altera Corporation
 
1 intro to_dpdk_and_hw
videos
 
5 pipeline arch_rationale
videos
 
6 profiling tools
videos
 
Sparc t4 systems customer presentation
solarisyougood
 
Revisit DCA, PCIe TPH and DDIO
Hisaki Ohara
 
The FlexTiles Development Platform offers Dual FPGA for 3D SoC Prototyping
FlexTiles Team
 
Hpe Proliant DL20 Gen10 Server Datasheet
美兰 曾
 
SNAPDRAGON SoC Family and ARM Architecture
Abdullaziz Tagawy
 
IGLOO2 Microsemi FPGA
Samira Riki
 
Rhino labs Prese4th ntation At FPGA Camp, Santa Clara, CA
FPGA Central
 
Session 8,9 PCI Express
Subhash Iyer
 
ODSA Use Case - SmartNIC
ODSA Workgroup
 
What are latest new features that DPDK brings into 2018?
Michelle Holley
 

Similar to Int 1010 Tcp Offload (20)

PDF
ARM LPC2300/LPC2400 TCP/IP Stack Porting
Mathivanan Elangovan
 
PDF
Ent voice ip-pbx-appliance
enterux_ltd
 
PPT
Modeling System Behaviors: A Better Paradigm on Prototyping
DVClub
 
PDF
Userspace networking
Stephen Hemminger
 
PDF
Xen summit 2010 extending xen into embedded
The Linux Foundation
 
PDF
In-Network Acceleration with FPGA (MEMO)
Naoto MATSUMOTO
 
PDF
Mellanox hpc day 2011 kiev
Volodymyr Saviak
 
PPT
Ethernet summit 2011_toe
intilop
 
PDF
509 512
Editor IJARCET
 
PDF
Big Data Smarter Networks
DataWorks Summit
 
PDF
Adv. FPGA Motor Control--EBV & Univ. of Koln: Embedded World 2010
Altera Corporation
 
PDF
数据中心网络研究:机遇与挑战
Weiwei Fang
 
PDF
A series presentation
Sergey Marunich
 
PDF
Benefits of Using FPGAs for Embedded Processing: Embedded World 2010
Altera Corporation
 
PPTX
Future Cloud Infrastructure
exponential-inc
 
PDF
XS Boston 2008 Network Topology
The Linux Foundation
 
PDF
Building Network Elements Using Intel Network Processors and ATCA
Joe Buzzanga
 
PDF
Flexible and Scalable Domain-Specific Architectures
Netronome
 
PPTX
Practica 2
Bruno Magallanes Noguez
 
PPTX
Design and Reuse Panel
Design And Reuse
 
ARM LPC2300/LPC2400 TCP/IP Stack Porting
Mathivanan Elangovan
 
Ent voice ip-pbx-appliance
enterux_ltd
 
Modeling System Behaviors: A Better Paradigm on Prototyping
DVClub
 
Userspace networking
Stephen Hemminger
 
Xen summit 2010 extending xen into embedded
The Linux Foundation
 
In-Network Acceleration with FPGA (MEMO)
Naoto MATSUMOTO
 
Mellanox hpc day 2011 kiev
Volodymyr Saviak
 
Ethernet summit 2011_toe
intilop
 
Big Data Smarter Networks
DataWorks Summit
 
Adv. FPGA Motor Control--EBV & Univ. of Koln: Embedded World 2010
Altera Corporation
 
数据中心网络研究:机遇与挑战
Weiwei Fang
 
A series presentation
Sergey Marunich
 
Benefits of Using FPGAs for Embedded Processing: Embedded World 2010
Altera Corporation
 
Future Cloud Infrastructure
exponential-inc
 
XS Boston 2008 Network Topology
The Linux Foundation
 
Building Network Elements Using Intel Network Processors and ATCA
Joe Buzzanga
 
Flexible and Scalable Domain-Specific Architectures
Netronome
 
Design and Reuse Panel
Design And Reuse
 
Ad

Int 1010 Tcp Offload

  • 1. TCP Offload Engine- SoftCore IP INT-1010 TCP Offload Engine Product brief, features and benefits summary Highly customizable hardware IP block. Easily portable to ASIC flow, Xilinx or Altera FPGAs INT-1010 is highly flexible that is customizable for layer-3, layer- 4, layer-5 network infrastructure and network security systems applications. It is recommended for use in, among others, high performance Servers, NICs, SAN/NAS and data center applications. It provides the key IP building block for a single high performance Giga bit Ethernet ASIC/ASSP/FPGA. INT-1010 provides capability for enterprises to differentiate their Network security and Network infrastructure appliances from others INT-1010 can process TCP/IP, for in-line network security appliances, sessions in both directions, simultaneously, at full G-bit rate. This relieves the host CPU from costly TCP/IP buffer stack execution and maintenance tasks. INT-1010 can process TCP/IP sessions and has the capability to process other relevant protocols such as FTP, UDP, ICMP, TFTP etc on all inbound as well as outbound traffic simultaneously without compromising performance at sustained Gigabit speeds. • Ideal for high performance • Scalable MAC Rx FIFOs and Tx and mid performance FIFOs make it ideal for specialized, differentiable optimizing system performance. ASICs or FPGAs for Network security or Network • Hardware implementation of infrastructure applications TCP/IP stacks’ control plane and data plane. • Less than 4000 Xilinx slices, Altera ALMs or 150,000 • Hardware implementation of ASIC gates + on-chip memory ARP protocol. • Fully integrated 100 M ARP table creation, deletion bit/1-G bit high performance management (optional) EMAC. Intelop Corporation www.intelop.com 4800 Great America Pkwy. Ste-231 Santa Clara, CA. 95054. Ph: 408-496-0333, Fax: 408-496-0444
  • 2. TCP Offload Engine- SoftCore IP • Customizable for IP-protocol live TCP sessions. only. Customizable, depending upon on-chip memory availability. • On-chip DDR or SSRAM memory controller which can address • Extendable to 256K TCP from 4K Bytes to 4 MB Bytes on sessions. Internal Memory chip or 256 MB off chip dependent. memories (User customizable, optional) • Wire-speed 2Gbps performance at full duplex • Many trade-offs for some functions performed in • Can process up to 256K hardware or software connections per second • Configurable Packet buffers, • TCP + IP check sum session table buffers On-chip generation and check performed or Off-chip memories, attached in hardware in less than 6 DDR I/II interface. Depending clks (30 ns at 200 MHz) vs 1-2 on system, performance, us by typical software TCP- ASIC/FPGA size requirements-> stack User Customizable, (optional) • Connection set up and tear • Interfaces directly to GMII, down/termination RGMII, MII external 10/100/ 1000 Mbit Phy interfaces • User programmable Session table parameters • Architecture can be scaled up to 10-G bits • Dedicated set of hardware Timers for each TCP/IP session • Customizable to handle jumbo or customizable for sharing frames stale sessions. • Generic bus interface for • Multiple ‘slot storage’ for Local Processor control. ARM, fragmented packets. More slots PPC, MIPS and other CPU allocated when more On-chip interfaces available Memory available. Self- checking available memory • User programmable/ logic. (optional) prioritize-able interrupts • Out of sequence packet • Performs connection/session detection/storage and management Reassembly/Segmentation (optional) • Monitors, Stores, Maintains and processes more than 4024 • RDMA- Direct Data placement in Applications buffer at full Intelop Corporation www.intelop.com 4800 Great America Pkwy. Ste-231 Santa Clara, CA. 95054. Ph: 408-496-0333, Fax: 408-496-0444
  • 3. TCP Offload Engine- SoftCore IP wire speed without CPU-> • Accommodates future reduces CPU’s buffer copy time Specifications changes. and utilization by 95% • Basic mini API available for • Future Proof- Flexible PPC architecture. Others implementation of TCP Offload OSs/CPUs also available. Simplified Block Diagram Intelop Corporation www.intelop.com 4800 Great America Pkwy. Ste-231 Santa Clara, CA. 95054. Ph: 408-496-0333, Fax: 408-496-0444
  • 4. TCP Offload Engine- SoftCore IP Specifications brief: • Functionality Proven in multiple IDS/IPS appliances • Complete header and flag processing of TCP/IP sessions in hardware accelerates by 10 x – 50 x • TCP Offload Engine- 2G b/s Wire-speed performance • Scalable to 10 G b/s • TCP + IP check sum- hardware • TCP segmentation/reassembly in hardware • Multiple ‘slot storage’ for fragmented packets • Out of sequence packet detection/storage/Reassembly • Accelerate security processing, Storage Networking- TCP • RDMA- Data placement in Applications buffer -> reduces CPU utilization by 90 % • Future Proof- Flexible implementation of TCP Offload • Accommodates future Specifications changes. • TOE MS-chimney architecture compatible • Customizable. Source code; Verilog • Detailed specs available under NDA Intelop Corporation www.intelop.com 4800 Great America Pkwy. Ste-231 Santa Clara, CA. 95054. Ph: 408-496-0333, Fax: 408-496-0444
  • 5. TCP Offload Engine- SoftCore IP Xilinx Altera iTOE - FPGA Altera Stratix III, IV, V Stratix - 2 /ALMs ~2256 ~3256 ~1256 16- 256 Alter-vqm, Verilog Verilog /Altera Quartus 8.1 Quartus 8.1 Intelop Corporation www.intelop.com 4800 Great America Pkwy. Ste-231 Santa Clara, CA. 95054. Ph: 408-496-0333, Fax: 408-496-0444