forked from docs/doc-exports
Wuwan, Qi
050b395397
Reviewed-by: Kacur, Michal <michal.kacur@t-systems.com> Co-authored-by: Wuwan, Qi <wuwanqi1@noreply.gitea.eco.tsi-dev.otc-service.com> Co-committed-by: Wuwan, Qi <wuwanqi1@noreply.gitea.eco.tsi-dev.otc-service.com>
27 lines
4.9 KiB
HTML
27 lines
4.9 KiB
HTML
<a name="css_04_0002"></a><a name="css_04_0002"></a>
|
|
|
|
<h1 class="topictitle1">Scenarios</h1>
|
|
<div id="body0000001426732772"><p id="css_04_0002__p158937442224">CSS can be used to build search boxes for websites and apps to improve user experience. You can also build a log analysis platform with it, facilitating data-driven O&M and business operations. CSS vector search can help you quickly build smart applications, such as AI-based image search, recommendation, and semantic search.</p>
|
|
<div class="section" id="css_04_0002__section560182917278"><h4 class="sectiontitle">Site Search</h4><p id="css_04_0002__p8614541101718">CSS can be used to search for website content by keyword as well as search for and recommend commodities on e-commerce sites.</p>
|
|
<ul id="css_04_0002__ul18460205211179"><li id="css_04_0002__li17460135201717">Real-time search: When site content is updated, you can find the updated content in your search within minutes, or even just seconds.</li><li id="css_04_0002__li1146035216174">Categorized statistics: You can apply search filters to sort products by category.</li><li id="css_04_0002__li246155216171">Custom highlight style: You can define how the search results are highlighted.</li></ul>
|
|
<div class="fignone" id="css_04_0002__fig15504173903016"><span class="figcap"><b>Figure 1 </b>Site search</span><br><span><img id="css_04_0002__image1450493933014" src="en-us_image_0000001715704521.png"></span></div>
|
|
</div>
|
|
<div class="section" id="css_04_0002__section11290124812317"><h4 class="sectiontitle">All-Scenario Log Analysis</h4><p id="css_04_0002__p7566185043314">Analyze the logs of Elastic Load Balance (ELB), servers, containers, and applications. In CSS, the Kafka message buffer queue is used to balance loads in peak and off-peak hours. Logstash is used for data extract, transform and load (ETL). Elasticsearch retrieves and analyzes data. The analysis results are visualized by Kibana and presented to you.</p>
|
|
<ul id="css_04_0002__ul55741333133419"><li id="css_04_0002__li3574113318342">High cost-effectiveness: CSS separates cold and hot storage, and decouples computing and storage resources, achieving high performance and reducing costs by over 30%.</li><li id="css_04_0002__li95745335346">Ease of use: Perform queries in a GUI editor. Easily create reports using drag-and-drop components.</li><li id="css_04_0002__li55741133113416">Powerful processing capability: CSS can import hundreds of terabytes of data per day, and can process petabytes of data.</li></ul>
|
|
<div class="fignone" id="css_04_0002__fig11794050153420"><span class="figcap"><b>Figure 2 </b>All-scenario log analysis</span><br><span><img id="css_04_0002__image87945505340" src="en-us_image_0000001667704906.png"></span></div>
|
|
</div>
|
|
<div class="section" id="css_04_0002__section16200161703510"><h4 class="sectiontitle">Database Query Acceleration</h4><p id="css_04_0002__p53501844134914">CSS can be used to accelerate database queries. E-commerce and logistics companies have to respond to a huge number of concurrent order queries within a short period of time. Relational databases, although having good transaction atomicity, are weak in transaction processing, and can rely on CSS to enhance OLTP and OLAP capabilities.</p>
|
|
<ul id="css_04_0002__ul1493361155717"><li id="css_04_0002__li793312112578">High performance: Retrieve data from hundreds of millions of records within milliseconds. Text, time, numeric, and spatial data types are supported.</li><li id="css_04_0002__li89331125718">High scalability: CSS can be scaled to have over 200 data nodes and over 1000 columns.</li><li id="css_04_0002__li79335116575">Zero service interruption: The rolling restart and dual-copy mechanisms can avoid service interruption in case of specifications change or configuration update.</li></ul>
|
|
</div>
|
|
<div class="section" id="css_04_0002__section9581045135715"><h4 class="sectiontitle">Vector Search</h4><p id="css_04_0002__p20383118175817">When you search for unstructured data, such as images, videos, and corpuses, the nearest neighbors or approximate nearest neighbors are searched based on feature vectors. This has the following advantages:</p>
|
|
<ul id="css_04_0002__ul1441916451405"><li id="css_04_0002__li542044512011">Efficient and reliable: The vector search engine provides optimal search performance and distributed DR capabilities.</li><li id="css_04_0002__li24203451908">Abundant indexes: Multiple indexing algorithms and similarity measurement methods are available and can meet diverse needs.</li><li id="css_04_0002__li204201845300">Easy learning: CSS is fully compatible with the open-source Elasticsearch ecosystem.</li></ul>
|
|
<div class="fignone" id="css_04_0002__fig815013782116"><span class="figcap"><b>Figure 3 </b>Vector search</span><br><span><img id="css_04_0002__image191501173213" src="en-us_image_0000001715624681.png"></span></div>
|
|
</div>
|
|
</div>
|
|
<div>
|
|
<div class="familylinks">
|
|
<div class="parentlink"><strong>Parent topic:</strong> <a href="css_01_0001.html">Overview</a></div>
|
|
</div>
|
|
</div>
|
|
|