Application Scenarios
IoT Device Monitoring
Application scenarios:
IoT devices such as IoE, gas, water, electricity, chemical devices, and Internet are connected to the cloud through IoT suite services. CloudTable keeps track of your devices, writes device data and analysis results to OpenTSDB in real time, and outputs time series results to the user's front-end monitoring system using OpenTSDB APIs. It implements real-time monitoring and analysis of IoT devices.
Advantages:
- Easy access
CloudTable (OpenTSDB) supports open protocols that enable easy interconnection with the messaging system and real-time stream computing system, reducing development difficulties.
- High-performance read/write
10-million-level write throughput of time series data, 3-second query latency for million-level data points; 30% to 60% read performance improvement and 60% write concurrency increase compared with open source OpenTSDB
- Aggregation capabilities
Interpolation, downsampling, and powerful aggregation function capabilities
- Low cost
On-demand storage charging and flexible capacity expansion helping cope with the uncertainty of services; high compression ratio (10:1), lowering costs
Related Services:
Cloud Stream Service (CS), Data Ingestion Service (DIS), Data Lake Insight (DLI), IoT Platform, and Object Storage Service (OBS)
Storage and Query of Message Logs
Application Scenarios:
Structured and semi-structured key-value data can be stored and queried, including messages, reports, recommendation data, risk control data, logs, and orders.
Advantages:
- Mass storage
Offline and online storage of massive volumes of key-value data, and flexible capacity expansion
- High-performance read/write
100-million-level write throughput, millisecond-level query latency for presenting online applications and reports
- Enriched ecosystem
Various Hadoop ecosystem components, integrated with HUAWEI CLOUD products
Related Services:
DIS and CS
Location-based Big Data Applications in IoV
Application Scenarios:
In IoV scenarios, various types of data are generated, such as basic information about vehicles and drivers; monitoring data of vehicle status, batteries, and motors; and vehicle driving data. GeoMesa stores and analyzes spatiotemporal data and provides functions such as tracking query, region distribution statistics, region query, density analysis, aggregation, and origin-destination (OD) analysis.
Advantages:
- Multimodal Database Capabilities
Different indexes provided for various types of data to deliver optimal performance, query, and analysis
- Abundant query and analysis functions
GeoMesa provides functions such as tracking query, region distribution statistics, region query, density analysis, aggregation, and OD analysis.
- Seamless interconnection
Spatiotemporal databases seamlessly interconnected with DLI to provide better analysis for time and space data, such as heat maps
Related Services:
CS, DIS, DLI, IoT Platform, and OBS
Profile Storage and Query
Application Scenarios:
Labels are used to describe characteristics of people and objects. Each person or object has a set of labels that are uncertain because data is frequently updated. This type of data is widely used in marketing decision-making, recommendation, and advertising systems.
Advantages
- Sparse matrix
The sparse matrix model of HBase is suitable for storing unstructured data. No schema needs to be predefined for tables and no strict column definition is required among rows.
- Update Anytime
You can update any rows at any time without performance loss. HBase itself versioning mechanism is used to save multiple historical versions of data.
Related Services:
DIS and CS
Serverless Web/Mobile App Back-ends
Application Scenarios:
CloudTable and FunctionGraph are collaborated to quickly build high available, auto-scaling web and mobile application back-ends.
Advantages:
- High availability
CloudTable and OBS HA ensuring high reliability of website data, and API Gateway and FunctionGraph improving HA of website logic
- Superb performance
A maximum of 20,000 IOPS per disk and 350 MB/s throughput
- Scalable
If services surge, resources can be automatically allocated to run more function instances to meet processing requirements.
- Low cost
You are only charged for the duration during which functions are processing files or data, and the storage capacity you use. Auto scaling enables you to avoid resource redundancy during off-peak hours.
Related Services:
OBS and FunctionGraph
Last Article: What Is CloudTable
Next Article: Cluster Mode Functions
Did this article solve your problem?
Thank you for your score！Your feedback would help us improve the website.