RE: You are right [was Re: Ungodly packet loss rates]
Another pragmatic solution is to call the editors of comm week, network world, data communications and suggest that they might get a lot of mileage writing a story comparing and contrasting the performance of ISPs. They do this for routers, bridges, FR services so they can probably find a respectable consulting/measurement group to collect information. Some of this could be live empirical data and some of this could be survey. cheers, peter
From Kent England
Unfortunately, the practical solution to your problem is to shop around and the more you know about your unique topological concerns the better off you'll be. I wish there was a Net Consumers Report service that you could subscribe to, but we aren't quite there yet.
So, John, relax a little knowing you are fundamentally right and your expectations are entirely reasonable and please accept my apology that the Internet isn't quite ready yet to meet your needs. I take solace in the fact that it has never been better and that it will continue to get better.
--Kent speaking as a long time Internet services person
Another pragmatic solution is to call the editors of comm week, network world, data communications and suggest that they might get a lot of mileage writing a story comparing and contrasting the performance of ISPs. They do this for routers, bridges, FR services so they can probably find a respectable consulting/measurement group to collect
Yes, they do. It would be better (sometimes) if they did not -:) When I read their comparations I sometimes think to drom all my hardware (BAD in terms of this magasines) and bue new one (Bay Networks as BB routers, FORE ATM as ATM , etc...). Through it's strange idea (for example) to compare hight-end Bay router with CS7200 (middle-range router), or to compare middle-range Lt1010 with Cascade back-bone ATM switchs. And I am afraid thy would use the same technik for ISP comparation. But anyway it's interesting idea... --- Aleksei Roudnev, Network Operations Center, Relcom, Moscow (+7 095) 194-19-95 (Network Operations Center Hot Line),(+7 095) 239-10-10, N 13729 (pager) (+7 095) 196-72-12 (Support), (+7 095) 194-33-28 (Fax)
ISPs. They do this for routers, bridges, FR services so they can probably find a respectable consulting/measurement group to collect Yes, they do. It would be better (sometimes) if they did not -:)
When I read their comparations I sometimes think to drom all my hardware (BAD in terms of this magasines) and bue new one (Bay Networks as BB routers, FORE ATM as ATM , etc...). Through it's strange idea (for example) to compare hight-end Bay router with CS7200 (middle-range router), or to compare middle-range Lt1010 with Cascade back-bone ATM switchs. And I am afraid thy would use the same technik for ISP comparation.
Well, besides being interesting ISP comparison *may* not be as objective and informative as we'd like to. In addition, what technique do you propose for such comparison? Just asking customers aren't enough, IMHO. And you can't count *every* ISP, even in the US... My 2c, Edgar -- Edgar V.S. Der-Danieliantz Armenia Network Information Center hostmaster@amnic.net Azatootianne 1, Yerevan, Armenia
Well, besides being interesting ISP comparison *may* not be as objective and informative as we'd like to. In addition, what technique do you propose for such comparison? Just asking customers aren't enough, IMHO. And you can't count *every* ISP, even in the US...
As for me, it is interesting two technoques used simultaneously: 1) Journal asks ISP to make dialup-IP and 64K account for the testing; and asks where they coud try T1 connection. 2) Journal bue (anonimously) dialup account from the same ISP's. Then, every day in 1 months (or some 5 different days during this months) they measure CPS for - WWW to some interesting pages, FTP from some servers, quality of real-audio connections, etc... Then (since 1 months) it will be very interestind data. But I am afrayd we'll see quite another picture - "we prefere XXX because they sell dialup-kit with MS Exploper, and do not recommend YYY because in our ONLY ftp test from www.cisco.com CPS was 5% less than in our 10 ftp tests via XXX_, or something simular.
My 2c,
Edgar
-- Edgar V.S. Der-Danieliantz Armenia Network Information Center hostmaster@amnic.net Azatootianne 1, Yerevan, Armenia
--- Aleksei Roudnev, Network Operations Center, Relcom, Moscow (+7 095) 194-19-95 (Network Operations Center Hot Line),(+7 095) 239-10-10, N 13729 (pager) (+7 095) 196-72-12 (Support), (+7 095) 194-33-28 (Fax)
As for me, it is interesting two technoques used simultaneously:
1) Journal asks ISP to make dialup-IP and 64K account for the testing; and asks where they coud try T1 connection.
2) Journal bue (anonimously) dialup account from the same ISP's.
Then, every day in 1 months (or some 5 different days during this months) they measure CPS for - WWW to some interesting pages, FTP from some servers, quality of real-audio connections, etc...
Then (since 1 months) it will be very interestind data.
Agreed. But you don't mention customer support, prices, services (many ISPs give you phone number, id/password - and you're on your way. others provide customized news, content, local information, support, etc...) In other words, there is a difference between "real" ISP and "not-so" ISP... -edd
Agreed too, I missed this. It means there have to be nessesary different rankes - quality of service, quality of support, prices, etc..., etc... It may be interesting even for us (there in Russia).
Agreed. But you don't mention customer support, prices, services (many ISPs give you phone number, id/password - and you're on your way. others provide customized news, content, local information, support, etc...)
In other words, there is a difference between "real" ISP and "not-so" ISP...
-edd
--- Aleksei Roudnev, Network Operations Center, Relcom, Moscow (+7 095) 194-19-95 (Network Operations Center Hot Line),(+7 095) 239-10-10, N 13729 (pager) (+7 095) 196-72-12 (Support), (+7 095) 194-33-28 (Fax)
On Wed, 23 Oct 96 14:23:15 +0400, alex@relcom.eu.net writes:
Another pragmatic solution is to call the editors of comm week, network world, data communications and suggest that they might get a lot of mileage writing a story comparing and contrasting the performance of ISPs. They do this for routers, bridges, FR services so they can probably find a respectable consulting/measurement group to collect
Yes, they do. It would be better (sometimes) if they did not -:)
When I read their comparations I sometimes think to drom all my hardware (BAD in terms of this magasines) and bue new one (Bay Networks as BB routers, FORE ATM as ATM , etc...). Through it's strange idea (for example) to compare hight-end Bay router with CS7200 (middle-range router), or to
You're probably thinking of the Network World review a few months back. As I understood it, Network World asked Cisco, Bay, and 3com to submit their high-end router for testing, and Cisco submitted the 7200. This had me a bit confused, too. Even with my limited Cisco knowledge, I know a 7200 isn't a high-end router. Someone told me that they probably submitted it because the 7200 had Netflow and the others didn't or something. Of course, Cisco dumped Netflow shortly after they came out with it, so I'm not sure what that tells you.. Why do you think it's bad that these reviews make you want to buy new hardware? Personally, I enjoy seeing what the competition is offering. Maybe by having Bay's BCN blow away the competition on packet throughput, it will encourage Cisco to get off their asses and build a scalable router that doesn't need to be replaced every year. And maybe it will encourage 3com to.. well... build a decent router period. :) -Jon ----------------------------------------------------------------- * Jon Green * Wide-Area Networking Technician * * jon@netINS.net * Iowa Network Services, Inc. * * Finger for Geek Code/PGP * 312 8th Street, Suite 730 * * #include "std_disclaimer.h" * Des Moines, IA 50309 * -------------------------------------------------------------------------
Why do you think it's bad that these reviews make you want to buy new hardware? Personally, I enjoy seeing what the competition is offering. Personally - because their results differs from my opinion totally - I'll never get some router because it's faster (interesting questions for me is - can it hold 40,000 routes? What's the cost of memory upgrade? How many HSSI, Serial and Ethernet interfaces I can plug in? Can I reconfigure BGP withouth reloading total router? and so on... No one answer. But - I get information _router XXX drops some packets if they try to cause it work with 10 FDDI links, etc..., etc... very interesting and absolutely useless...
And when I ask some network administrator about this, I get direct answer - router XXX have not _telnet_ configuration option, router YYY have not 'ip classless', router ZZZ use 85% of CPU when driving 16 Async links on 115,200 bps - its' an answer for me. But their comparation... brr. Another example - modems. I can't speak for USA, but there (in Russia) modems differ by their stability, interoperability, LL options, _if I can restrict lower speed of modem connection_, _if I can cause modem don't try too hight rate in case of bad line_, etc, etc... No one answer I can found in the review (sorry, I forget when I read it last) - but I read _modem XXX is faster than YYY on 20%_ - very interesting...
Maybe by having Bay's BCN blow away the competition on packet throughput, it will encourage Cisco to get off their asses and build a scalable router that doesn't need to be replaced every year. And maybe it will encourage 3com to.. well... build a decent router period. :)
Aleksei Roudnev, Network Operations Center, Relcom, Moscow (+7 095) 194-19-95 (Network Operations Center Hot Line),(+7 095) 239-10-10, N 13729 (pager) (+7 095) 196-72-12 (Support), (+7 095) 194-33-28 (Fax)
On Wed, 23 Oct 96 17:01:27 +0400, alex@relcom.eu.net writes:
Why do you think it's bad that these reviews make you want to buy new hardware? Personally, I enjoy seeing what the competition is offering. Personally - because their results differs from my opinion totally - I'll never get some router because it's faster (interesting questions for me is - can it hold 40,000 routes? What's the cost of memory upgrade? How many HSSI, Serial and Ethernet interfaces I can plug in? Can I reconfigure BGP withouth reloading total router? and so on... No one answer. But - I get information _router XXX drops some packets if they try to cause it work with 10 FDDI links, etc..., etc... very interesting and absolutely useless...
I wouldn't say it's absolutely useless. Your router sales rep or SE can tell you the features of the router, and you can verify those features yourself. The point where I stop necessarily believing my sales rep is when it comes to performance, and that's when independent side by side testing is useful. As for speed, it may not be an issue with you- yet. Anyone running an OC-3 Internet backbone should be *very* concerned about speed, though. They should also be concerned about scalability. Performance is not the only factor to consider when buying a router (and you'll note the Network World article considered a lot more than just performance), but it is definitely something to consider. Remember, a lot of the current scaling problems the Internet is having is because people use routers that look pretty and have nice software features, but have dismal hardware. "It can hold 40,000 routes, but if we get up to 50,000 I'll have to buy a new router because this one can't hold any more memory." In short, when spending the money for a high-end router, NO information is useless. -Jon ----------------------------------------------------------------- * Jon Green * Wide-Area Networking Technician * * jon@netINS.net * Iowa Network Services, Inc. * * Finger for Geek Code/PGP * 312 8th Street, Suite 730 * * #include "std_disclaimer.h" * Des Moines, IA 50309 * -------------------------------------------------------------------------
I am not shure this is for nanog, through:
No one answer. But - I get information _router XXX drops some packets if they try to cause it work with 10 FDDI links, etc..., etc... very interesting and absolutely useless... ... it is definitely something to consider. Remember, a lot of the current scaling problems the Internet is having is because people use routers that look pretty and have nice software features, but have dismal hardware. "It can hold 40,000 routes, but if we get up to 50,000 I'll have to buy a new router because this one can't hold any more memory." You are right. But unfortunately this means the Cisco is the worst selection because their hardware is badly scaled (let's compare stackable Bay router, new Ascend's router and any - any - Cisco), they cause customer to bue new CS4700 instead of CS4500 to make simple memory upgrade (isn't it amazing? any Pentium PC can use more than 64Mb RAM and hard, solid Cisco 4500 /with MIPS processor etc.../ can't? Is it good hardware) from 32 to 64Mb RAM. And there is a lot of such examples when Cisco's solution looks badky via bad hardware.
May be it's Cisco's problem; may be it's the play of Cisco's sales, bay be it's reality - I do not know. But they cause small and middla-range ISP to choose nonscalable solutions. And then total Internet society loss total quality via this choose. This moves me to the old idea - it's good news for ISP if any hardware vendor makes new, good and competive router. Back to the _big brother_ subject - most ISP likes CISCO, and most hate it at the same time.
In short, when spending the money for a high-end router, NO information is useless.
For old, solid network administrators - yes. But I see every day how some brain-less manager in some company choose XXX-vendor's router because he have read excellent article about this router - and then we (as ISP) spend our time trieing this router to work (yes, we get our reward - but we are ISP, not _bad hardware consultants_). --- Aleksei Roudnev, Network Operations Center, Relcom, Moscow (+7 095) 194-19-95 (Network Operations Center Hot Line),(+7 095) 239-10-10, N 13729 (pager) (+7 095) 196-72-12 (Support), (+7 095) 194-33-28 (Fax)
well if you're going to compare ciscos and bay networks routers, consider that Bay networks supports Rip, OSPF, BGP, and EGP. They do *NOT* support communities in their production software, and they have *NO* intentions of *EVER* supporting confederations. In adition, to handle subnets, where you want the thing to summarise a subnet into a classful route, the Bay's solution is to drop the route entirely. They also don't seem to understand how to aggregate routes. Their solution there is also to drop the route. They do not appear to have the option to announce the aggregate with the routes. They also do not appear to have the option of aggregating since the option they provide does not work. Their SNMP agent only works on a few platforms, and in order to adequately solve a routing problem, you need to have a *GOOD* understanding of the MIB. The last time I enabled syslog on the box, the router reloaded several times within a 5 hour period, causing instability in our small network, small meaning under 200 routes. I have fought with these things for 3 years now and haven't seen much improvements. They have been promising NTP support for quite some time now, since their routers don't have a battery-powered clock. Maybe the reason they can switch packets faster and more reliably than ciscos is because they are unable to be placed in a situation to really test their skills. The items I have shown here make it VERRY difficult to allow one of these things to perform with full routing because you cannot determine what it will do. -- ------------------------------------------- | Jeremy Hall Network Engineer | | ISDN-Net, Inc Office +1-615-371-1625 | | Nashville, TN and the southeast USA | | jhall@isdn.net Pager +1-615-702-0750 | -------------------------------------------
Jeremy, The statements below contain so many inaccuracies that I am not even going to try to correct them... I think this kind of needless and completely inaccurate bashing is *way* out of line... -Marten "Mr. Jeremy Hall" <jhall@rex.isdn.net> writes * well if you're going to compare ciscos and bay networks routers, consider * that Bay networks supports Rip, OSPF, BGP, and EGP. They do *NOT* * support communities in their production software, and they have *NO* * intentions of *EVER* supporting confederations. In adition, to handle * subnets, where you want the thing to summarise a subnet into a classful * route, the Bay's solution is to drop the route entirely. They also don't * seem to understand how to aggregate routes. Their solution there is also * to drop the route. They do not appear to have the option to announce the * aggregate with the routes. They also do not appear to have the option of * aggregating since the option they provide does not work. Their SNMP * agent only works on a few platforms, and in order to adequately solve a * routing problem, you need to have a *GOOD* understanding of the MIB. The * last time I enabled syslog on the box, the router reloaded several times * within a 5 hour period, causing instability in our small network, small * meaning under 200 routes. I have fought with these things for 3 years * now and haven't seen much improvements. They have been promising NTP * support for quite some time now, since their routers don't have a * battery-powered clock. Maybe the reason they can switch packets faster * and more reliably than ciscos is because they are unable to be placed in * a situation to really test their skills. The items I have shown here * make it VERRY difficult to allow one of these things to perform with * full routing because you cannot determine what it will do. * -- * ------------------------------------------- * | Jeremy Hall Network Engineer | * | ISDN-Net, Inc Office +1-615-371-1625 | * | Nashville, TN and the southeast USA | * | jhall@isdn.net Pager +1-615-702-0750 | * ------------------------------------------- *
I can't say about Bay routers, but I know one system admin who was caused (by him manager) to build IP/IPX network on Bay routers. He hate Bay and dreams about Cisco. It's terrible... Yes, it looks fine, but it's terrible device. Configuration, reloading, etc.,.. - good hardware and bad software (yet). But _nanog_ is not place for this flame - everybody may have him opinion. And (even if somebody hate Bay's) there is a lot of cases when Bay is better than other routers.
Jeremy,
The statements below contain so many inaccuracies that I am not even going to try to correct them... I think this kind of needless and completely inaccurate bashing is *way* out of line...
-Marten
"Mr. Jeremy Hall" <jhall@rex.isdn.net> writes
--- Aleksei Roudnev, Network Operations Center, Relcom, Moscow (+7 095) 194-19-95 (Network Operations Center Hot Line),(+7 095) 239-10-10, N 13729 (pager) (+7 095) 196-72-12 (Support), (+7 095) 194-33-28 (Fax)
At 7:28 -0500 96/10/23, Jon Green wrote:
On Wed, 23 Oct 96 14:23:15 +0400, alex@relcom.eu.net writes:
Another pragmatic solution is to call the editors of comm week, network world, data communications and suggest that they might get a lot of mileage writing a story comparing and contrasting the performance of ISPs. They do this for routers, bridges, FR services so they can probably find a respectable consulting/measurement group to collect
Yes, they do. It would be better (sometimes) if they did not -:)
When I read their comparations I sometimes think to drom all my hardware (BAD in terms of this magasines) and bue new one (Bay Networks as BB routers, FORE ATM as ATM , etc...). Through it's strange idea (for example) to compare hight-end Bay router with CS7200 (middle-range router), or to
You're probably thinking of the Network World review a few months back. As I understood it, Network World asked Cisco, Bay, and 3com to submit their high-end router for testing, and Cisco submitted the 7200. This had me a bit confused, too. Even with my limited Cisco knowledge, I know a 7200 isn't a high-end router. Someone told me that they probably submitted it because the 7200 had Netflow and the others didn't or something. Of course, Cisco dumped Netflow shortly after they came out with it, so I'm not sure what that tells you..
The 7500 series supports NetFlow. The ISP's currently testing/using netflow in their networks would be quite surprised to learn that cisco has "dumped" it. The development engineers working on further improving its usefulness, both for switching and for gathering of statistics would also be astonished ;-) Robert.
Why do you think it's bad that these reviews make you want to buy new hardware? Personally, I enjoy seeing what the competition is offering. Maybe by having Bay's BCN blow away the competition on packet throughput, it will encourage Cisco to get off their asses and build a scalable router that doesn't need to be replaced every year. And maybe it will encourage 3com to.. well... build a decent router period. :)
-Jon
----------------------------------------------------------------- * Jon Green * Wide-Area Networking Technician * * jon@netINS.net * Iowa Network Services, Inc. * * Finger for Geek Code/PGP * 312 8th Street, Suite 730 * * #include "std_disclaimer.h" * Des Moines, IA 50309 * -------------------------------------------------------------------------
participants (7)
-
alex＠relcom.eu.net -
edd＠acm.org -
Jon Green -
Marten Terpstra -
Mr. Jeremy Hall -
Peter Ford -
Robert Craig