Solution Performance

Baosight xInsight* is an industrial big data platform which is deployed in manufacturing, transportation and government fields. It is used for storage, query and analysis of big data. Major customers include Bao Steel* and Chongqing (Shanghai) Metro*.

Visit Baosight.com ›

Challenges

Customers experience an I/O bottleneck due to limited main memory today, which underutilizes the CPU and limits throughput that can be achieved. Adding additional memory is often cost prohibitive, especially for large deployments.

Solution

  • Baosight xInsight* application is limited by available memory, resulting in an I/O bottleneck. Intel® Optane™ DC persistent memory (App Direct mode) delivers significantly larger, affordable memory capacities, as well as the flexibility to allocate data structures to specific memory tiers.
  • This allows customers to have larger datasets in memory, therefore delivering better performance with more transactions per second (see chart for performance improvement).1 2 3

Value

Better performance at similar cost - Baosight customers will benefit from improved performance (i.e., being able to run more transactions per second), at comparable cost, and while meeting SLA target of a response time less than 10ms.

產品與效能資訊

1

效能結果依據 Intel 與 Baosight 於 2019 年 1 月 8 日的測試,可能無法反映所有公開提供的安全性更新。沒有產品或元件能提供絕對的安全性。

2

Shanghai Baosight xInsight* v2.0(自定義工作負載);作業系統:CentOS* 7.5 核心 3.10.0-957.1.3.el7.x86_64。由 Intel 和 Baosight 在 2019 年 1 月 8 日所完成的測試。針對變異 1、2、3 與 L1TF 的安全性風險緩解已準備就緒。 

基準:第 2 代 Intel® Xeon® Platinum 處理器 8260L,2.3 GHz,24 核心,開啟渦輪與超執行緒,BIOS 1.0180,總記憶體 768GB,32 插槽 / 32GB / 2666 MT/s / DDR4 LRDIMM,1 x 480GB / Intel® SSD Data Center (Intel® SSD DC) S4500 + 2 x 1TB / Intel® SSD DC P4500。

新組態:第 2 代 Intel® Xeon® Platinum 處理器 8260L,2.3 GHz,24 核心,開啟渦輪與超執行緒,BIOS 1.0180,總記憶體 192GB,12 插槽 / 16GB / 2666 MT/s / DDR4 LRDIMM 與 1024GB DCPMM,8 插槽 / 128 GB / 2666 MT/s Intel® Optane™ DC 持續性記憶體,1 x 480GB / Intel® SSD DC S4500 + 2 x 1TB / Intel® SSD DC P4500。

3

效能測試中使用的軟體與工作負載可能僅針對 Intel® 微處理器進行最佳化。包括 SYSmark* 與 MobileMark* 在內的效能測試是使用特定電腦系統、零組件、軟體、作業與功能進行量測。這些因素若有任何異動,均可能導致測得結果產生變化。建議您參考其他資訊與效能測試數據,協助您充分評估欲購買產品的性能,包括該產品在搭配其他產品運作時的效能。如需詳細資訊,請參閱 www.intel.com.tw/benchmarks

4

IBM Aspera FASP NX*. Workload drives memory-to-memory transfer of data between 2 systems directly connected via 3x 40Gbe Ethernet links. Testing by Intel and IBM. April-August 2017.
Baseline: 2S Intel® Xeon® processor E5-2650 v4, 2.2GHz, 12 cores, turbo, HT on, BIOS 0271.R00, 256GB total memory, 8 slots / 32GB / 2400 MT/s / DDR4 LRDIMM, CentOS* 7.3 kernel 3.10.0-514.16.1.el7.x86_64, 3x Intel® XL710. Source: Aspera testing as of August 2017. Measured effective throughput: 15Gbps per core. Equivalent to downloading ~52 30X compressed whole human genomes per core per hour.
New: 2S Intel® Xeon® Platinum 8168 processor, 2.7GHz, 24 cores, turbo on, HT on, BIOS 00.0412, 192GB total memory, 12 slots / 16GB / 2667 MT/s / DDR4 LRDIMM, 1 x 800GB Intel® SSD DC S3700, 3x Intel® XL710 10/40Gbe Ethernet Controller, Red Hat Enterprise Linux* 7.3 kernel 3.10.0-514.16.1.el7.x86_64. Source: Aspera testing as of August 2017. Measured effective throughput: 23Gbps per core. Equivalent to downloading ~80 30X whole human genomes per core per hour.
Assumptions: one whole human genome, raw reads, 30x coverage including quality scores, 130GB compressed, source: https://medium.com/precision-medicine/how-big-is-the-human-genome-e90caa3409b0.

5

效能標竿結果是在實施最近的軟體修補程式與韌體更新以解決「Spectre」和「Meltdown」安全漏洞之前取得的資料。實施這些更新可能會讓這些結果變得不適用於您的裝置或系統。

效能測試中使用的軟體與工作負載可能僅針對 Intel® 微處理器進行最佳化。包括 SYSmark* 與 MobileMark* 在內的效能測試是使用特定電腦系統、零組件、軟體、作業與功能進行量測。這些因素若有任何異動,均可能導致測得結果產生變化。建議您參考其他資訊與效能測試數據,協助您充分評估欲購買產品的性能,包括該產品在搭配其他產品運作時的效能。如需完整的資訊,請參閱 http://www.intel.com.tw/benchmarks