Clip from Cisco.com As technology and features mature, they often move from a software-based implementation to inclusion in hardware. At the core of Cisco's hardware integration is application-specific integrated circuit (ASIC) innovation. Cisco has developed more than one hundred ASICs for the Cisco Catalyst switching family over the past nine years, with each generation including more capabilities. For example, Cisco was the first vendor to integrate Layer 3 switching into hardware with the Cisco Catalyst 5500 NetFlow Feature Card. With the introduction of the Cisco Catalyst 6500 Series in 1999, Cisco included features such as quality of service (QoS) classification and queuing and security access control lists (ACLs), and provided them at data rates of millions of packets per second. These features are available, in hardware, across the Cisco Catalyst switching product line, including the Cisco Catalyst 4500, 3750, and 3560. Advanced hardware integration continues with the Cisco Catalyst 6500 Series Supervisor Engine 720, which integrates MPLS, IPv6, and generic routing encapsulation (GRE). This is the first time a LAN switch has offered this capability at data rates in the hundreds of millions of packets per second.
Yes most routing is done in software, especially if there are policies to apply. Typically, however, this is just the first packet of a conversation and the balance of the traffic is switched in hardware by referencing tables, aka MLS, router on a stick, route once switch many, etc...
Luca Deri's work was done for use in packet filtering but that does not mean it cannot be applied to this topic. In order to route a packet it first has to get to the kernel. In this manner the two functions are identical (BPF filter aside).
Lastly it is not the hardware that is the problem but a buffer and kernel one. The problem is that the kernel gets an interrupt for every frame that the NIC receives, gets inundated with them and drops packets. Luca's article illustrates this fairly clearly.
There is NO way software routing can compare to processing packets in hardware. The Linux kernel wasn't designed for this and has problems when faced with a large number of packets. I'll reference the work done by Luca Deri at NTOP.org and his pfring mod. Unless we start seeing specialized open source hardware I don't think Cisco will feel threatened in the least.
Once upon a time it was very easy for me to find answers to my questions online. Most web content now is sadly much more interested with separating you from your cash than sharing information. Librarians are slowly embracing technology and mastering the (art?) of digital storage. Once they do, I'm hopeful they will be able to help restore some of the Internet's content.
I would love to subscribe to a number of scientific jornals but at >$200/year there's no way I could justify it. I understand these are small distribution publications that don't have the economy of scale that say newspapers enjoy. Although the material they print is donated (correct me if I'm wrong on this), publication & distribution is expensive with little commercial space to offset the cost. By using electronic distribution maybe the prices can come down to the level at which your average Joe could afford them.
Why not have 2 seats per flight reserved for members of the Armed Forces who, after passing a competency and skill evaluation process (ie mastery of the Vulcan death pinch), can fly for free if they agree to "work" that flight.
The Fed, when they put their minds to it, can come up with some really stupid things. A result, primarily, of scores of clueless gov't employees/officials that must justify their positions by contributing something. This can be anything, effective or counter-effective, logical or just plan cockamamie, it's all the same.
Too bad two of the largest companies, with the most resources to contest SCO's claim, have decided to roll over and play dead. I guess it's up to the rest of us to fight it. Anyone have a link for a group spear heading this effort?
It's somewhat assured nowadays that a smart bomb will hit it's intended target and that any incidental loss/casualties will be localized. With cyber terrorism suddenly physical location doesn't matter as much. Could illegal hacking be masked by the campaign?
Oh, your website selling 'Peace in Iraq' bumber stickers was hacked? Gee sorry, our bad. We thought you were a military target.
Encryption is a rat race and complete security is undeniably impossible. A best effort method should be employed that makes recording and playback about as difficult as normal cell transmission interception. Possible but out of the reach of the majority. -jeff
I'm ok with my voice being xmitted over the carrier's network but this is a little disconcerting. What degree of privacy is afforded by a random public access point operated by some random individual? Isn't WiFi really a shared medium? think:WarEavesdropping. I sincerely hope this concern is being addressed. -Jeff
Where were we on optical technology 5 years ago? Can you imagine where we'll be 5 from today? DWDM will be superceeded and OC-768's & OC-3072's will be commonplace. Technology will provide for squeezing more data into the same # of physical fibers. Have faith.
I just had my one-year anniversary as an IT contractor for the US Census Bureau. Although my job description reads otherwise I am the unofficial lead NOC Engineer (can say the e word since I have by BS in eng). What I've discovered for the most part is that the gov't employees are grossly under qualified and often down right stupid. The Gov't hires from within and often promotes people who don't have a clue and never will. Since the authority & responsibility is spread so thin, people can easily hide from accountability by finger pointing.
Now the contractors are supposed to introduce knowledge, talent and intelligence and do most of the work but defer the credit to the gov't workers. For example: as a new hire I was chastised when I would write memos like, "When I discovered..." Instead I'm supposed to include the group and use "we" because even though no one else did anything they still deserve credit. This unspoken rule naturally doesn't apply to the gov't workers.
Anyhow my experience as a Federal contractor has been an education in the unbelievable inefficiency of the Government. I'm currently looking hard for new employment in the private sector. Here I'm just getting a paycheck. Zero growth and zero satisfaction.
Slashdot Mirror
ASIC= Application Specific Integrated Circuit, and yea, Cisco's stuff is chock full of em.
Router# show mls asic
Cafe version: 2
Centauri version: 1
Perseus version: 0/0
Titan version: 1
Clip from Cisco.com
As technology and features mature, they often move from a software-based implementation to inclusion in hardware. At the core of Cisco's hardware integration is application-specific integrated circuit (ASIC) innovation. Cisco has developed more than one hundred ASICs for the Cisco Catalyst switching family over the past nine years, with each generation including more capabilities. For example, Cisco was the first vendor to integrate Layer 3 switching into hardware with the Cisco Catalyst 5500 NetFlow Feature Card. With the introduction of the Cisco Catalyst 6500 Series in 1999, Cisco included features such as quality of service (QoS) classification and queuing and security access control lists (ACLs), and provided them at data rates of millions of packets per second. These features are available, in hardware, across the Cisco Catalyst switching product line, including the Cisco Catalyst 4500, 3750, and 3560. Advanced hardware integration continues with the Cisco Catalyst 6500 Series Supervisor Engine 720, which integrates MPLS, IPv6, and generic routing encapsulation (GRE). This is the first time a LAN switch has offered this capability at data rates in the hundreds of millions of packets per second.
Your flaming isn't so good.
Yes most routing is done in software, especially if there are policies to apply. Typically, however, this is just the first packet of a conversation and the balance of the traffic is switched in hardware by referencing tables, aka MLS, router on a stick, route once switch many, etc...
Luca Deri's work was done for use in packet filtering but that does not mean it cannot be applied to this topic. In order to route a packet it first has to get to the kernel. In this manner the two functions are identical (BPF filter aside).
Lastly it is not the hardware that is the problem but a buffer and kernel one. The problem is that the kernel gets an interrupt for every frame that the NIC receives, gets inundated with them and drops packets. Luca's article illustrates this fairly clearly.
There is NO way software routing can compare to processing packets in hardware. The Linux kernel wasn't designed for this and has problems when faced with a large number of packets. I'll reference the work done by Luca Deri at NTOP.org and his pfring mod. Unless we start seeing specialized open source hardware I don't think Cisco will feel threatened in the least.
Akamai gets the credit for that.
Once upon a time it was very easy for me to find answers to my questions online. Most web content now is sadly much more interested with separating you from your cash than sharing information. Librarians are slowly embracing technology and mastering the (art?) of digital storage. Once they do, I'm hopeful they will be able to help restore some of the Internet's content.
I would love to subscribe to a number of scientific jornals but at >$200/year there's no way I could justify it. I understand these are small distribution publications that don't have the economy of scale that say newspapers enjoy. Although the material they print is donated (correct me if I'm wrong on this), publication & distribution is expensive with little commercial space to offset the cost. By using electronic distribution maybe the prices can come down to the level at which your average Joe could afford them.
The content is rather small with only 1878 entries. The ICAT database, however, is mature with 6548 entries.
...to have to remake in 2-3 years.
Why are manufacturers still using WEP? Oh, wait, could it be so companies who NEED it pay for the higher priced models?
A step in the right direction but I'm afraid the problem is too international to be affected much.
Why not have 2 seats per flight reserved for members of the Armed Forces who, after passing a competency and skill evaluation process (ie mastery of the Vulcan death pinch), can fly for free if they agree to "work" that flight.
The Fed, when they put their minds to it, can come up with some really stupid things. A result, primarily, of scores of clueless gov't employees/officials that must justify their positions by contributing something. This can be anything, effective or counter-effective, logical or just plan cockamamie, it's all the same.
Too bad two of the largest companies, with the most resources to contest SCO's claim, have decided to roll over and play dead. I guess it's up to the rest of us to fight it. Anyone have a link for a group spear heading this effort?
It's somewhat assured nowadays that a smart bomb will hit it's intended target and that any incidental loss/casualties will be localized. With cyber terrorism suddenly physical location doesn't matter as much. Could illegal hacking be masked by the campaign?
Oh, your website selling 'Peace in Iraq' bumber stickers was hacked? Gee sorry, our bad. We thought you were a military target.
Encryption is a rat race and complete security is undeniably impossible. A best effort method should be employed that makes recording and playback about as difficult as normal cell transmission interception. Possible but out of the reach of the majority.
-jeff
I'm ok with my voice being xmitted over the carrier's network but this is a little disconcerting. What degree of privacy is afforded by a random public access point operated by some random individual? Isn't WiFi really a shared medium? think:WarEavesdropping. I sincerely hope this concern is being addressed.
-Jeff
Where were we on optical technology 5 years ago? Can you imagine where we'll be 5 from today? DWDM will be superceeded and OC-768's & OC-3072's will be commonplace. Technology will provide for squeezing more data into the same # of physical fibers. Have faith.
A little dated but still contains usefull insight.
Linux IDE-RAID Notes
I just had my one-year anniversary as an IT contractor for the US Census Bureau. Although my job description reads otherwise I am the unofficial lead NOC Engineer (can say the e word since I have by BS in eng). What I've discovered for the most part is that the gov't employees are grossly under qualified and often down right stupid. The Gov't hires from within and often promotes people who don't have a clue and never will. Since the authority & responsibility is spread so thin, people can easily hide from accountability by finger pointing.
Now the contractors are supposed to introduce knowledge, talent and intelligence and do most of the work but defer the credit to the gov't workers. For example: as a new hire I was chastised when I would write memos like, "When I discovered..." Instead I'm supposed to include the group and use "we" because even though no one else did anything they still deserve credit. This unspoken rule naturally doesn't apply to the gov't workers.
Anyhow my experience as a Federal contractor has been an education in the unbelievable inefficiency of the Government. I'm currently looking hard for new employment in the private sector. Here I'm just getting a paycheck. Zero growth and zero satisfaction.