Search Results

Search found 37 results on 2 pages for 'jchang'.

Page 1/2 | 1 2  | Next Page >

• Hyper-Threading comments

  - by jchang

  There seems to be significant confusion about Hyper-Threading. Part of the problem is that vendors like to tout every new feature as the greatest invention since the six-pack, and its follow-on the 12-pack. I used to think the 4-pack was a travesty, but now that I am older and can nolonger finish a 12-pack with each meal, suddenly the 4-pack is not such a travesty. But I digress. I do appluad innovation, and I do accept that the first generation is almost never perfect, thats why its the bleeding...(read more)

  Read the article

• Gigabit and Full-Duplex

  - by jchang

  People still talk about checking if the network is in full-duplex mode even when they are on Gigabit Ethernet. Let me say clearly: Gigabit Ethernet is full-duplex period. There is no half-duplex mode. The same goes for 10 Gigabit Ethernet. If Windows Task Manager says the network Link Speed is 1 or 10 Gbps, don’t bother checking the mode, it can only be full-duplex. In the old days of 10Mbit/sec Ethernet was originally half-duplex. The old 10BASE5 (fat) and 10BASE2 (thin) cable had one signal carrier....(read more)

  Read the article

• Intel Xeon 5600 (Westmere-EP) and AMD Magny-Cours Performance Update

  - by jchang

  HP has just released TPC-C and TPC-E results for the ProLiant DL380G7 with 2 Xeon 5680 3.33GHz 6-core processor, allowing a direct comparison with their DL385G& with 2 Opteron 6176 2.3GHz 12-core processors. Last month I complained about the lack of performance results for the Intel Xeon 5600 6-core 32nm processor line for 2-way systems. This might have been deliberate to not complicate the message for the Xeon 7500 8-core 45nm (for 4-way+ systems) launch two weeks later. http://sqlblog.com/blogs/joe_chang/archive/2010/04/07/intel-xeon-5600-westmere-ep-and-7500-nehalem-ex.aspx...(read more)

  Read the article

• Storage Configuration

  - by jchang

  Storage performance is not inherently complicated subject. The concepts are relatively simple. In fact, scaling storage performance is far easier compared with the difficulties encounters in scaling processor performance in NUMA systems. Storage performance is achieved by properly distributing IO over: 1) multiple independent PCI-E ports (system memory and IO bandwith is key) 2) multiple RAID controllers or host bus adapters (HBAs) 3) multiple storage IO channels (SAS or FC, complete path) most importantly,...(read more)

  Read the article

• Storage Configuration

  - by jchang

  Storage performance is not inherently complicated subject. The concepts are relatively simple. In fact, scaling storage performance is far easier compared with the difficulties encounters in scaling processor performance in NUMA systems. Storage performance is achieved by properly distributing IO over: 1) multiple independent PCI-E ports (system memory and IO bandwith is key) 2) multiple RAID controllers or host bus adapters (HBAs) 3) multiple storage IO channels (SAS or FC, complete path) most importantly,...(read more)

  Read the article

• Intel Xeon 5600 (Westmere-EP) and 7500 (Nehalem-EX)

  - by jchang

  Intel Xeon 5600 (Westmere-EP) and 7500 (Nehalem-EX) Performance Intel launched the Xeon 5600 series (Westmere-EP, 32nm) six-core processors on 16 March 2010 without any TPC benchmark results. In the performance world, no results almost always mean bad or not good results. Yet there is every reason to believe that the Xeon 5600 series with six-cores (X models only) will performance exactly as expected for a 50% increase in the number of cores at the same frequency (as the 5500) with no system level...(read more)

  Read the article

• IBM System x3850 X5 TPC-H Benchmark

  - by jchang

  IBM just published a TPC-H SF 1000 result for their x3850 X5 , 4-way Xeon 7560 system featuring a special MAX5 memory expansion board to support 1.5TB memory. In Dec 2010, IBM also published a TPC-H SF1000 for their Power 780 system, 8-way, quad-core, (4 logical processors per physical core). In Feb 2011, Ingres published a TPC-H SF 100 on a 2-way Xeon 5680 for their VectorWise column-store engine (plus enhancements for memory architecture, SIMD and compression). The figure table below shows TPC-H...(read more)

  Read the article

• Data, Log and Temp file placement

  - by jchang

  First especially for all the people with SAN storage, drive letters are of no consequence. What matters is the actual physical disk layout. Forget capacity, pay attention to the number of spindles supporting each RAID group. If the RAID group is shared with other application, make sure there the SLA guarantees read and write latency. One very large company conducted a stress test in the QA environment. The SAN admin carved the LUNs from the same pool of disks as production, but thought he had a really...(read more)

  Read the article

• HP ProLiant DL980-Oracle TPC-C Benchmark spat

  - by jchang

  The Register reported a spat between HP and Oracle on the TPC-C benchmark. Per above, HP submitted a TPC-C result of 3,388,535 tpm-C for their ProLiant DL980 G7 (8 Xeon X7560 processors), with a cost of $0.63 per tpm-C. Oracle has refused permission to publish. Late last year (2010) Oracle published a result of 30M tpm-C for a 108 processors (sockets) SPARC cluster ($30M complete system cost). Oracle is now comparing this to the HP Superdome result from 2007 of 4M tpm-C at $2.93 per tpm-C, calling...(read more)

  Read the article

• Automating SQL Execution Plan analysis

  - by jchang

  Last year, I made my tool for automating execution plan analysis available on www.qdpma.com The original version could parse execution plans from sys.dm_exec_query_stats or dm_exec_cached_plans and generate a cross-reference of which execution plans employed each index. The DMV sys.dm_db_index_usage_stats shows how often each index is used, but not where, that is, which particular stored procedure or My latest version can now also 1) use the DMV sys.dm_exec_procedure_stats, 2) it can also get the...(read more)

  Read the article

• Intel Xeon E5 (Sandy Bridge-EP) and SQL Server 2012 Benchmarks

  - by jchang

  Intel officially announced the Xeon E5 2600 series processor based on Sandy Bridge-EP variant with upto 8 cores and 20MB LLC per socket. Only one TPC benchmark accompanied product launch, summary below. Processors Cores per Frequency Memory SQL Vendor TPC-E 2 x Xeon E5-2690 8 2.9GHz 512GB (16x32GB) 2012 IBM 1,863.23 2 x Xeon E7-2870 10 2.4GHz 512GB (32x16GB) 2008R2 IBM 1,560.70 2 x Xeon X5690 6 3.46GHz 192GB (12x16GB) 2008R2 HP 1,284.14 Note: the HP report lists SQL Server 2008 R2 Enterprise Edition...(read more)

  Read the article

• Server Systems for SQL Server 2012 per core licensing

  - by jchang

  Until recently, the SQL Server Enterprise Edition per processor (socket) licensing model resulted in only 2 or 3 server system configurations being the preferred choice. Determine the number of sockets: 2, 4 or 8. Then select the processor with the most compute capability at that socket count level. Finally, fill the DIMM sockets with the largest capacity ECC memory module at reasonable cost per GB. Currently this is the 16GB DIMM with a price of $365 on the Dell website, and $240 from Crucial. The...(read more)

  Read the article

• Dell whitepaper on PowerEdge R810 R910 and M910 Memory Architecture

  - by jchang

  The Dell PowerEdge 11 th Generation Servers: R810, R910 and M910 Memory Guidance whitepaper seems to have caused some confusion. I believe the source is an error in the paper. In the section on FlexMem Bridge Technology, the Dell whitepaper says this applies to the R810 and the M910. The Dell M910 is a 4-way blade server for the Xeon 7500 series processor line. First a breif recap. The R810 is a 2-way server, by which I mean it has two sockets regardless of the number of cores on each processor....(read more)

  Read the article

• Columnar Databases

  - by jchang

  Ingres just published a TPC-H benchmark for VectorWise , an analytic database technology employing 1) SIMD processing (Intel SSE 4.2), 2) better memory optimizations to leverage on-chip cache, 3) compression, 4) Column-based storage. Ingres originated as a research project at UC Berkeley (see Wikipedia ) in the 1970s, and has since become a commercially supported, open source database system. Apparently, Ingres project people later founded Sybase. So Ingres in a sense, is the grandfather (or perhap...(read more)

  Read the article

• Intel Server Strategy Shift with Sandy Bridge EN & EP

  - by jchang

  The arrival of the Sandy Bridge EN and EP processors, expected in early 2012, will mark the completion of a significant shift in Intel server strategy. For the longest time 1995-2009, the strategy had been to focus on producing a premium processor designed for 4-way systems that might also be used in 8-way systems and higher. The objective for 2-way systems was use the desktop processor that later had a separate brand and different package & socket to leverage the low cost structure in driving...(read more)

  Read the article

• Oracle Index Skip Scan

  - by jchang

  There is a feature, called index skip scan that has been in Oracle since version 9i. When I across this, it seemed like a very clever trick, but not a critical capability. More recently, I have been advocating DW on SSD in approrpiate situations, and I am thinking this is now a valuable feature in keeping the number of nonclustered indexes to a minimum. Briefly, suppose we have an index with key columns: Col1 , Col2 , in that order. Obviously, a query with a search argument (SARG) on Col1 can use...(read more)

  Read the article

• Parallel Data Warehouse

  - by jchang

  The Microsoft Parallel Data Warehouse diagram was somewhat difficult to understand in terms of the functionality of each subsystem in relation to the configuration of its components. So now that HP has provided a detailed list of the PDW components , the diagram below shows the PDW subsystems with component configuration (InfiniBand, FC, and network connections not shown). Observe that there are three different ProLiant server models, the DL360 G7, DL370 G6 and the DL380 G7, in five different configurations...(read more)

  Read the article

• SIMD Extensions for the Database Storage Engine

  - by jchang

  For the last 15 years, Intel and AMD have been progressively adding special purpose extensions to their processor architectures. The extensions mostly pertain to vector operations with Single Instruction, Multiple Data (SIMD) concept. The reasoning was that achieving significant performance improvement over each successive generation for the general purpose elements had become extraordinarily difficult. On the other hand, SIMD performance could be significantly improved with special purpose registers...(read more)

  Read the article

• New SQL Server 2012 per core licensing – Thank you Microsoft

  - by jchang

  Many of us have probably seen the new SQL Server 2012 per core licensing, with Enterprise Edition at $6,874 per core super ceding the $27,495 per socket of SQL Server 2008 R2 (discounted to $19,188 for 4-way and $23,370 for 2-way in TPC benchmark reports) with Software Assurance at $6,874 per processor? Datacenter was $57,498 per processor, so the new per-core licensing puts 2012 EE on par with 2008R2 DC, at 8-cores per socket. This is a significant increase for EE licensing on the 2-way Xeon 5600...(read more)

  Read the article

• SSD-HDD price parity

  - by jchang

  It is hard to believe that we are essentially at SSD-HDD price parity? Of course I am comparing enterprise class 10K/15K HDDs to consumer grade SSDs. Below are prices I am seeing 300GB 15K HDD $370 900GB 10K HDD $600 1TB 7200 HDD $230 (less for consumer HDDs) 512GB SATA SSD $400-600 Intel SSD DC S3700 400GB $940 The 512GB SATA SSDs are consumer grade, MLC NAND, with only 7% over provisioning. That is 512GB (1GB = 2^30) of NAND, with 512GB (1GB =10^9) of user capacity. Intel just announced the SSD...(read more)

  Read the article

• Free tools for SQL Server - Automating Execution Plan Analysis

  - by jchang

  Since this topic is being discussed, I will plug my own tools, SQL Exec Stats and (a little dated) documentation the main capability is cross-referencing index usuage with specific execution plans. another feature is generating execution plans for all stored procedures in a database, along with the index usage cross-reference. There are several sources of execution plans or plan handles, this could be a live trace, a previously saved trace, previously saved sqlplan files, from dm_exec_cached_plans,...(read more)

  Read the article

• Fast Track Data Warehouse 3.0 Reference Guide

  - by jchang

  Microsoft just release Fast Track Data Warehouse 3.0 Reference Guide version. The new changes are increased memory recommendation and the disks per RAID group change from 2-disk RAID 1 to 4-Disk RAID 10. Memory The earlier FTDW reference architecture cited 4GB memory per core. There was no rational behind this, but it was felt some rule was better than no rule. The new FTDW RG correctly cites the rational that more memory helps keep hash join intermediate results and sort operations in memory. 4-Disk...(read more)

  Read the article

• Backup Compression - time for an overhaul

  - by jchang

  Database backup compression is incredibly useful and valuable. This became popular with then Imceda (later Quest and now Dell) LiteSpeed. SQL Server version 2008 added backup compression for Enterprise Edition only. The SQL Server EE native backup feature only allows a single compression algorithm, one that elects for CPU efficiency over the degree of compression achieved. In the long ago past, this strategy was essential. But today the benefits are irrelevant while the lower compression is becoming...(read more)

  Read the article

• SIMD Extensions for the Database Storage Engine

  - by jchang

  For the last 15 years, Intel and AMD have been progressively adding special purpose extensions to their processor architectures. The extensions mostly pertain to vector operations with Single Instruction, Multiple Data (SIMD) concept. The motivation was that achieving significant performance improvement over each successive generation for the general purpose elements had become extraordinarily difficult. On the other hand, SIMD performance could be significantly improved with special purpose registers...(read more)

  Read the article

• Brute Force Hardware versus Tuning

  - by jchang

  Every now and then, the question is asked “When will servers be powerful enough that performance tuning will not be necessary.” Sometimes I get the impression this was posed not on technical grounds, but rather that ISVs and CIOs don’t like performance specialists. Fortunately (for me) it does not appear that this will ever happen for two principal reasons: 1) hardware and tuning contribute in such completely different aspects that neither can be ignored, and 2) multi-core processors actually introduce...(read more)

  Read the article

1 2  | Next Page >