I have also had good results with the TP-LINK TL-WR1043ND Ultimate Wireless N Gigabit Router running OpenWrt and use them as network firewall/gateways, OpenVPN server/clients and small VLAN-capable switches.
The hardware specs are great for the price ($55-$65): 5 gigabit ports, 802.11n and the real winner is that OpenWrt has good support for the RTL8366RB switch chipset, including tagged VLANs. Getting a switch that has VLAN support for this price is not otherwise available, that I've found. The stock TP-LINK firmware does not expose the VLAN features but they are there and available if OpenWrt is used.
- Chipset is the Atheros AR9132 rev. 2 WISOC, which contains: Atheros 3×2 (3rx, 2 tx) MIMO wireless + MIPS 24Kc V7.4 CPU running at 400Mhz) – 266.24 BogoMIPS - RAM: 32 MB - FLASH NVRAM: 8 MB - 801.11n/g/b 2.4 GHz - 4+1 gigabit Ethernet ports (RTL8366RB chipset with 802.1Q VLAN support – see http://www.realtek.com.tw/products/productsView.aspx?Langid=1&PNid=18&PFid=15&Level=5&Conn=4&ProdID=197) - 1 USB 2.0 (can put a small USB stick in and use it as swap space to run programs that otherwise wouldn't work given 32 MB of RAM, such as dansguardian)
The one main problem I've found is that some wireless clients have problems connecting using 802.11 -- see the bug "Backfire ar71xx error: ath: Failed to stop TX DMA!" at https://dev.openwrt.org/ticket/9693 for examples. This has improved with -rc5 and -rc6.
One thing that burned our test users with Impress is that when images are dragged-and-dropped in, they are linked, not embedded (PowerPoint defaults to embedding). This caused one of our users to take an image-heavy presentation they had carefully assembled and have no images show up on the presentation machine.
When you add images using the menu in Impress (testing with 3.0), you can choose to link or embed, but when dragging-and-dropping, which is what our users were used to, the default is link. This cannot be changed as far as I could determine (see bug 15369).
Impress also converts embedded images to PNG, which can really inflate their size compared to JPG.
To give some context, this is a short column I wrote for this week's (4/21/2003) eWEEK news package on Windows Server 2003. It's short because of print space limitations. The whole collection of related news articles in this week's issue is at http://www.eweek.com/category2/0,3960,1034194,00.a sp.
Next week, eWEEK is publishing an eWEEK Labs review of the product. In that package, there are six pages of copy covering Windows Server 2003 overall security changes, IIS 6.0, 64-bit Windows, Active Directory changes, file and print changes, development, and storage and SAN changes.
Thanks,
Tim Dyck
eWEEK Labs West Coast Technical Director
Hi Spoing, the story doesn't position Tux as a competitor to Apache. In fact, we went out of our way to test the combination of Apache and Tux working together, as well as Tux and Apache (and IIS) on their own. We point out how well Tux and Apache work together and recommend that combination.
You may have come to your conclusion only on the basis of the title of the Web article, which is different than the print version. I think the print title is better, which is Tux: Built for Speed.
Also, I think that painting all stories from all Ziff-Davis publications with the same brush is too broad a generalization. The company produces content aimed at everything from home users and gamers to IT managers at companies that spend millions of dollars a year on technology (the later is eWEEK's market). You're more likely to find stories you like by following the work of particular authors or publications than the activities of an entire publisher.
Regards,
Tim Dyck
West Coast Technical Director
eWEEK Labs
It was available all along on eWEEK's site, along with an Excel spreadsheet giving more detailed test configurations and the exact pages/sec numbers. ZDNet has the right to republish eWEEK content and didn't include the page with the associated downloads in their version. The page is http://www.zd net.com/eweek/stories/general/0,11011,2646732,00.html. Unfortunately, there is are some spaces in the name of the ZIP file that has the code: Mozilla and IE download the file fine, but Netscape 4.7 has a problem with the filename. I'll get that changed, and can also e-mail it to anyone who wants it.
I found out a few days ago that this story got posted on Slashdot and will post a response to address some of the questions that have come up as soon as I can get it done.
Regards,
Tim Dyck, eWEEK Labs
timothy_dyck@ziffdavis.com
We will publish this if we can repo it and it will affect enough people -- and we will keep your identity anonymous. You can even e-mail me anonymously if you want to go though a remailer (e.g. http://anon.xg.nu or https://www.privacyx.com).
Regards, Tim Dyck Technical Director, eWEEK Labs timothy_dyck@ziffdavis.com
Hi Chris, my statement is based on what is currently shipping as InterBase 5.6 and what I know about PG 7. Of course, neither IB 6 or PG 7 has shipped yet, and so we don't know for sure what will be in the final bits. For PostgreSQL, having declarative referential integrity and outer joins (when they make it in, whether in 7.0 or 7.1) will be big steps forward.
In any case, it's great that there will be two transactional open source databases availble, and I'm sure each will push the other in good ways. More choice is a great thing.
I don't think you will find database failover in an open source product -- it's too specialized an area to have gathered interest from the necessary critical mass of developers.
I cover databases for PC Week Labs, and in my view (and in my tests), the most sophisticated OSS relational database is PostgreSQL. In three months, though, that title will be held by InterBase. It has a more complete SQL implementation, has a (third-party) replication option and is generally more mature. However, it doesn't have any failover.
The products that do provide failover are (non exhaustive list, but these are the main ones): Oracle Parallel Server (or Oracle with the Failsafe option if you want just 2 node failover), Sybase ASE, IBM DB2 UDB, Informix, Tandem NonStop SQL and Microsoft SQL Server. The cheapest option in terms of purchase price will be Microsoft SQL Server with Microsoft Cluster Server, but it only runs on NT and only supports 2 node failover, and does not cluster.
Note that when you talk about failover (particularly with stateful connections), you need to also get failover-capable hardware systems. The databases mentioned above all need special, tightly spec'ed high availability hardware configurations from Sun, HP, etc. In general, a shared SCSI bus is required.
You also need to decide what level of failover you need: (from simplest and cheapest to most complex, expensive and best):
* cold standby (replicate to a standby server) -- no shared storage, mostly up-to-date, cheap, manual intervention required for failover
* warm standby (shared storage or transactional replication to a standby server) -- ensures no data loss for all committed transactions, requires fault-tolerant hardware and a fault-tolerant dbms, failover is automatic but may take several minutes to roll the log forward, warm up the cache and reconnect users
* hot standby (shared storage) -- no data loss for committed transactions, fast failover (30 seconds or less) -- this is a very specialized area, and you should have your dbms vendor work with you on setting up these kinds of HA setups. This is an ongoing area of research; Sybase 12 and Oracle8i 2.0 introduced new features specifically in this area to do things like pre-connect users and pre-warm the cache to speed failover times.
If you want to have an OSS solution, I'd advise using PostgreSQL or InterBase and writing stored procedures in C to replicate inserts, updates and deletes to a second server, then coding reconnect and heartbeat logic into your front-end apps. This will be easiest to do with an app server since you have a single point of data access. This will not provide atomicity or durability though (unless the updates use 2PC) -- you'll need to run a consistency check on the db on failover.
This is hairy, though. Overall, I advise paying for something that someone else has gotten working.
The only two packages I know about that support JSPs are IBM's WebSphere Studio and Macromedia's Drumbeat 2000. WebSphere is a higher end tool and Drumbeat is more of a visual designer for people who don't want to work at the code level. Both are a few hundred dollars as far as I recall. WebSphere Studio (when used with WebSphere) is unique in that it can debug JSPs in original source form, which is a real bit of wizardry becuase all JSPs must be run through the servlet precompiler before they can be run.
I get PR reps or vendors calling occasionally too, asking if they can place copy in the magazine. Duh, hello! It's pretty fun telling them to go find another publication.
Certainly at PC Week (and I think at most publications), there is a huge wall between edit and advertising. I don't even know who works in our ad department, I never get phone calls or e-mails from them, and I never know what ads are going to be in the magazine. I don't even really look, actually. The only material affect ads have on my day-to-day life is they affect how many pages of edit there are, since the ad-to-edit ratio is mandated by the postal service (to get a particular mailing rate). When we have more ads in a given week, we need to write more, and vice versa.
People charging that ad dollars affect type or amount of coverage are being sucked in by a seductive argument, but one that just isn't true. Other issues, like unfamiliarity with technology x or the effectiveness of the PR company representing company y are examples of factors that actually do make a difference.
If you want to build a custom Web-based OLAP client, I'd check out Internetivity's dbProbe (www.dbprobe.com) and AlphaBlox's AlphaBlox (www.alphablox.com).
dbProbe stands out because of its very small Java client (about 100KB the last time I tested it), while still providing a full-featured interface similar to the fat client products (Cognos, Brio, BusinessObjects, etc.). Its cube is generated as ASCII text, so it's pretty malliable. The server runs on Linux or you can go directly to a bigger back-end like MetaCube or TM1. InterNetivity did lots of things right here.
AlphaBlox is a newer and generally rawer product, but is specifically designed as a toolkit for building custom OLAP and reporting applications. Using JavaScript, HTML and their Java components, you can assemble a vertical OLAP application in a few hours. It's all client-side, so your cubes need to be relatively small (3 or 4 dimensions).
There are lots of other Web-based interfaces (virtually every OLAP vendor has one now), but the rest are quite closed and intended to be used largely as-is.
Hi all, I cover databases for PC Week Labs and have followed this story personally since Ellison made his $1 million comment. As others have noted, Oracle8's new (at the time) support for materialized views was absolutely key to Oracle's position. Oracle carried out this TPC-D test using a beta version of Oracle8 and were chomping at the bit to rub Microsoft's nose in some of their new benchmark results.
What materialized views allowed Oracle to do is *pre-calculate* results for the TPC-D queries so when they were actually clocked, the database optimizer just had to realize that pre-computed results tables were already available and simply retrieve the results. Materialized views are very useful in the real world, but they, of course, just shift computation time from query execution back to during the load and index stage. You need to be aware of the pros and cons of the approach.
Despite Oracle getting such a PR win over this, preaggregation is not their invention: IBM developed the technique (which it calls materialized views) and shipped it to the market before Oracle did. Oracle was real carefull not to allow DB2 to participate in their $1 million challenge!
Microsoft has, in fact, met this challenge, in that they computed the results of TPC-D query 5 in better than 100X Oracle's time (actually, Microsoft query 5 run times were around 1 second). In this test, they used their OLAP Services Component of SQL Server 7, which uses precalculation just as Oracle's materialized views do -- something I think is fair. However, the version 2.1 TPC-D rules state: "The TPC-D database must be implemented using a commercially available database management system (DBMS) and the queries executed via an interface using dynamic SQL" (page 6). Basically, the TPC is a relational SQL benchmarking organization, and you have to use a relational database product in its tests. OLAP Services is NOT a relational database and it does not have a SQL interface: it is a multidimensional database and uses an interface Microsoft developed called OLE DB for OLAP. Thus, Microsoft's OLAP Services results were not eligible for TPC-D auditing, and without query pre-calculation in SQL Server 7 itself, there was no way possible Microsoft could do query 5 fast enough to win. This, of course, is exactly what Oracle was counting on.:)
Microsoft's Data Transformation Services component moves only data between two database engines. Metadata like relational integrity constraints and uniqueness constraints are not transferred, nor are triggers or any procedural code. The one exception is in the case of users moving tables between two SQL Server installations, in which case data integrity constraints are also transferred (trigger and stored procedure code can also be transferred in this case, but not through DTS itself).
I have to say this is the funniest damn thing I have read in a WHILE! I went upstairs to take a leak and was still laughing to hard I pissed all over the toilet. It was worth it, though.
Slashdot Mirror
I have also had good results with the TP-LINK TL-WR1043ND Ultimate Wireless N Gigabit Router running OpenWrt and use them as network firewall/gateways, OpenVPN server/clients and small VLAN-capable switches.
The hardware specs are great for the price ($55-$65): 5 gigabit ports, 802.11n and the real winner is that OpenWrt has good support for the RTL8366RB switch chipset, including tagged VLANs. Getting a switch that has VLAN support for this price is not otherwise available, that I've found. The stock TP-LINK firmware does not expose the VLAN features but they are there and available if OpenWrt is used.
- Chipset is the Atheros AR9132 rev. 2 WISOC, which contains: Atheros 3×2 (3rx, 2 tx) MIMO wireless + MIPS 24Kc V7.4 CPU running at 400Mhz) – 266.24 BogoMIPS
- RAM: 32 MB
- FLASH NVRAM: 8 MB
- 801.11n/g/b 2.4 GHz
- 4+1 gigabit Ethernet ports (RTL8366RB chipset with 802.1Q VLAN support – see http://www.realtek.com.tw/products/productsView.aspx?Langid=1&PNid=18&PFid=15&Level=5&Conn=4&ProdID=197)
- 1 USB 2.0 (can put a small USB stick in and use it as swap space to run programs that otherwise wouldn't work given 32 MB of RAM, such as dansguardian)
The one main problem I've found is that some wireless clients have problems connecting using 802.11 -- see the bug "Backfire ar71xx error: ath: Failed to stop TX DMA!" at https://dev.openwrt.org/ticket/9693 for examples. This has improved with -rc5 and -rc6.
-Tim Miller Dyck
One thing that burned our test users with Impress is that when images are dragged-and-dropped in, they are linked, not embedded (PowerPoint defaults to embedding). This caused one of our users to take an image-heavy presentation they had carefully assembled and have no images show up on the presentation machine.
When you add images using the menu in Impress (testing with 3.0), you can choose to link or embed, but when dragging-and-dropping, which is what our users were used to, the default is link. This cannot be changed as far as I could determine (see bug 15369).
Impress also converts embedded images to PNG, which can really inflate their size compared to JPG.
To give some context, this is a short column I wrote for this week's (4/21/2003) eWEEK news package on Windows Server 2003. It's short because of print space limitations. The whole collection of related news articles in this week's issue is at http://www.eweek.com/category2/0,3960,1034194,00.a sp.
Next week, eWEEK is publishing an eWEEK Labs review of the product. In that package, there are six pages of copy covering Windows Server 2003 overall security changes, IIS 6.0, 64-bit Windows, Active Directory changes, file and print changes, development, and storage and SAN changes.
Thanks,
Tim Dyck
eWEEK Labs West Coast Technical Director
Hi, this story is just one part of a larger package published in this week's eWEEK. In that story, we tested Tux, Tux with Apache, Apache and IIS. The link is http://www.zdnet.com/eweek/stories/general/0,11011 ,2774242,00.html.
Regards,
Tim Dyck
eWEEK Labs
Hi Spoing, the story doesn't position Tux as a competitor to Apache. In fact, we went out of our way to test the combination of Apache and Tux working together, as well as Tux and Apache (and IIS) on their own. We point out how well Tux and Apache work together and recommend that combination.
You may have come to your conclusion only on the basis of the title of the Web article, which is different than the print version. I think the print title is better, which is Tux: Built for Speed.
Also, I think that painting all stories from all Ziff-Davis publications with the same brush is too broad a generalization. The company produces content aimed at everything from home users and gamers to IT managers at companies that spend millions of dollars a year on technology (the later is eWEEK's market). You're more likely to find stories you like by following the work of particular authors or publications than the activities of an entire publisher.
Regards,
Tim Dyck
West Coast Technical Director
eWEEK Labs
It was available all along on eWEEK's site, along with an Excel spreadsheet giving more detailed test configurations and the exact pages/sec numbers. ZDNet has the right to republish eWEEK content and didn't include the page with the associated downloads in their version. The page is http://www.zd net .com/eweek/stories/general/0,11011,2646732,00.html . Unfortunately, there is are some spaces in the name of the ZIP file that has the code: Mozilla and IE download the file fine, but Netscape 4.7 has a problem with the filename. I'll get that changed, and can also e-mail it to anyone who wants it.
I found out a few days ago that this story got posted on Slashdot and will post a response to address some of the questions that have come up as soon as I can get it done.
Regards,
Tim Dyck, eWEEK Labs
timothy_dyck@ziffdavis.com
We will publish this if we can repo it and it will affect enough people -- and we will keep your identity anonymous. You can even e-mail me anonymously if you want to go though a remailer (e.g. http://anon.xg.nu or https://www.privacyx.com).
Regards,
Tim Dyck
Technical Director, eWEEK Labs
timothy_dyck@ziffdavis.com
Hi Chris, my statement is based on what is currently shipping as InterBase 5.6 and what I know about PG 7. Of course, neither IB 6 or PG 7 has shipped yet, and so we don't know for sure what will be in the final bits. For PostgreSQL, having declarative referential integrity and outer joins (when they make it in, whether in 7.0 or 7.1) will be big steps forward.
In any case, it's great that there will be two transactional open source databases availble, and I'm sure each will push the other in good ways. More choice is a great thing.
-Tim Dyck
I don't think you will find database failover in an open source product -- it's too specialized an area to have gathered interest from the necessary critical mass of developers.
I cover databases for PC Week Labs, and in my view (and in my tests), the most sophisticated OSS relational database is PostgreSQL. In three months, though, that title will be held by InterBase. It has a more complete SQL implementation, has a (third-party) replication option and is generally more mature. However, it doesn't have any failover.
The products that do provide failover are (non exhaustive list, but these are the main ones): Oracle Parallel Server (or Oracle with the Failsafe option if you want just 2 node failover), Sybase ASE, IBM DB2 UDB, Informix, Tandem NonStop SQL and Microsoft SQL Server. The cheapest option in terms of purchase price will be Microsoft SQL Server with Microsoft Cluster Server, but it only runs on NT and only supports 2 node failover, and does not cluster.
Note that when you talk about failover (particularly with stateful connections), you need to also get failover-capable hardware systems. The databases mentioned above all need special, tightly spec'ed high availability hardware configurations from Sun, HP, etc. In general, a shared SCSI bus is required.
You also need to decide what level of failover you need: (from simplest and cheapest to most complex, expensive and best):
* cold standby (replicate to a standby server) -- no shared storage, mostly up-to-date, cheap, manual intervention required for failover
* warm standby (shared storage or transactional replication to a standby server) -- ensures no data loss for all committed transactions, requires fault-tolerant hardware and a fault-tolerant dbms, failover is automatic but may take several minutes to roll the log forward, warm up the cache and reconnect users
* hot standby (shared storage) -- no data loss for committed transactions, fast failover (30 seconds or less) -- this is a very specialized area, and you should have your dbms vendor work with you on setting up these kinds of HA setups. This is an ongoing area of research; Sybase 12 and Oracle8i 2.0 introduced new features specifically in this area to do things like pre-connect users and pre-warm the cache to speed failover times.
If you want to have an OSS solution, I'd advise using PostgreSQL or InterBase and writing stored procedures in C to replicate inserts, updates and deletes to a second server, then coding reconnect and heartbeat logic into your front-end apps. This will be easiest to do with an app server since you have a single point of data access. This will not provide atomicity or durability though (unless the updates use 2PC) -- you'll need to run a consistency check on the db on failover.
This is hairy, though. Overall, I advise paying for something that someone else has gotten working.
Regards,
Tim Dyck
Senior Analyst
PC Week Labs
The only two packages I know about that support JSPs are IBM's WebSphere Studio and Macromedia's Drumbeat 2000. WebSphere is a higher end tool and Drumbeat is more of a visual designer for people who don't want to work at the code level. Both are a few hundred dollars as far as I recall. WebSphere Studio (when used with WebSphere) is unique in that it can debug JSPs in original source form, which is a real bit of wizardry becuase all JSPs must be run through the servlet precompiler before they can be run.
-Tim Dyck, PC Week Labs
I get PR reps or vendors calling occasionally too, asking if they can place copy in the magazine. Duh, hello! It's pretty fun telling them to go find another publication.
Certainly at PC Week (and I think at most publications), there is a huge wall between edit and advertising. I don't even know who works in our ad department, I never get phone calls or e-mails from them, and I never know what ads are going to be in the magazine. I don't even really look, actually. The only material affect ads have on my day-to-day life is they affect how many pages of edit there are, since the ad-to-edit ratio is mandated by the postal service (to get a particular mailing rate). When we have more ads in a given week, we need to write more, and vice versa.
People charging that ad dollars affect type or amount of coverage are being sucked in by a seductive argument, but one that just isn't true. Other issues, like unfamiliarity with technology x or the effectiveness of the PR company representing company y are examples of factors that actually do make a difference.
- Tim Dyck, Senior Analyst, PC Week Labs
If you want to build a custom Web-based OLAP client, I'd check out Internetivity's dbProbe (www.dbprobe.com) and AlphaBlox's AlphaBlox (www.alphablox.com).
dbProbe stands out because of its very small Java client (about 100KB the last time I tested it), while still providing a full-featured interface similar to the fat client products (Cognos, Brio, BusinessObjects, etc.). Its cube is generated as ASCII text, so it's pretty malliable. The server runs on Linux or you can go directly to a bigger back-end like MetaCube or TM1. InterNetivity did lots of things right here.
AlphaBlox is a newer and generally rawer product, but is specifically designed as a toolkit for building custom OLAP and reporting applications. Using JavaScript, HTML and their Java components, you can assemble a vertical OLAP application in a few hours. It's all client-side, so your cubes need to be relatively small (3 or 4 dimensions).
There are lots of other Web-based interfaces (virtually every OLAP vendor has one now), but the rest are quite closed and intended to be used largely as-is.
Regards,
Tim Dyck
PC Week Labs
Hi all, I cover databases for PC Week Labs and have followed this story personally since Ellison made his $1 million comment. As others have noted, Oracle8's new (at the time) support for materialized views was absolutely key to Oracle's position. Oracle carried out this TPC-D test using a beta version of Oracle8 and were chomping at the bit to rub Microsoft's nose in some of their new benchmark results.
:)
What materialized views allowed Oracle to do is *pre-calculate* results for the TPC-D queries so when they were actually clocked, the database optimizer just had to realize that pre-computed results tables were already available and simply retrieve the results. Materialized views are very useful in the real world, but they, of course, just shift computation time from query execution back to during the load and index stage. You need to be aware of the pros and cons of the approach.
Despite Oracle getting such a PR win over this, preaggregation is not their invention: IBM developed the technique (which it calls materialized views) and shipped it to the market before Oracle did. Oracle was real carefull not to allow DB2 to participate in their $1 million challenge!
Microsoft has, in fact, met this challenge, in that they computed the results of TPC-D query 5 in better than 100X Oracle's time (actually, Microsoft query 5 run times were around 1 second). In this test, they used their OLAP Services Component of SQL Server 7, which uses precalculation just as Oracle's materialized views do -- something I think is fair.
However, the version 2.1 TPC-D rules state: "The TPC-D database must be implemented using a commercially available database management system (DBMS) and the queries executed via an interface using dynamic SQL" (page 6). Basically, the TPC is a relational SQL benchmarking organization, and you have to use a relational database product in its tests. OLAP Services is NOT a relational database and it does not have a SQL interface: it is a multidimensional database and uses an interface Microsoft developed called OLE DB for OLAP. Thus, Microsoft's OLAP Services results were not eligible for TPC-D auditing, and without query pre-calculation in SQL Server 7 itself, there was no way possible Microsoft could do query 5 fast enough to win. This, of course, is exactly what Oracle was counting on.
Regards,
Tim Dyck
PC Week Labs
Microsoft's Data Transformation Services component moves only data between two database engines. Metadata like relational integrity constraints and uniqueness constraints are not transferred, nor are triggers or any procedural code. The one exception is in the case of users moving tables between two SQL Server installations, in which case data integrity constraints are also transferred (trigger and stored procedure code can also be transferred in this case, but not through DTS itself).
Regards,
Tim Dyck
PC Week Labs
I have to say this is the funniest damn thing I have read in a WHILE! I went upstairs to take a leak and was still laughing to hard I pissed all over the toilet. It was worth it, though.
Thanks Don!