Sign in Subscribe

Topic

Global

Discover PAASUP's products, news, and trends for international audiences. Stay informed with our latest developments, company milestones, and insights in data intelligence and platform innovation.

CDC Implementation from PostgreSQL to StarRocks: A Practical Guide

This case study demonstrates building a real-time CDC pipeline from PostgreSQL to StarRocks using Debezium and Kafka, achieving 4,000+ rows/sec replication. With StarRocks' MPP architecture, we significantly improved analytical query performance and implemented a scalable data pipeline.

DuckDB: A Modern Analytics Database Guide

DuckDB is an embedded analytical database designed as SQLite's analytics counterpart. It enables fast analysis of large datasets locally without server setup and directly queries CSV, Parquet, and JSON files. This guide covers DuckDB's architecture, installation, and practical SQL examples.

PAASUP Selected as Top 15 Finalists for 'NVIDIA Inception Startup Grand Challenge 2025'

PAASUP Stands Shoulder-to-Shoulder with Leading AI Companies, Recognized for Private AI Innovation Technology

Extending Notebook-Centric AI Workflows with Jupyter AI

This article introduces how to extend your Jupyter Notebook environment into a generative AI playground using Jupyter AI. From installation methods to chat interface usage, key slash (/) commands, and %%ai magic commands, we cover the essential features step by step.

No-ETL Data Pipeline Using StarRocks MViews

Discover how to integrate diverse data sources in real-time using StarRocks Materialized Views without complex ETL. This guide covers complete pipeline implementation from data loading to visualization, featuring StarRocks and Apache Iceberg scenarios for cost reduction and high performance.

Deploy StarRocks on Kubernetes

This technical blog provides a hands-on guide for deploying StarRocks distributed analytics database on Kubernetes, covering both Shared-nothing and Shared-data architectures with Helm Charts, MinIO integration, and real-world performance testing to help choose the right deployment mode.

Understanding StarRocks Architecture

StarRocks is an analytical database with a simple FE/BE architecture offering two modes: Shared-nothing (local storage, high performance) and Shared-data (object storage, cost-efficient). It provides excellent MPP performance and seamless horizontal scaling without external dependencies.

Building a Text Curation Pipeline with NeMo Curator

This guide uses NVIDIA NeMo Curator to build high-quality datasets for LLMs. It's a hands-on walk-through of a text curation pipeline—from cleaning and deduplication to language labeling—for preparing large-scale data quickly and reliably.

Evaluating LLMs with NeMo Evaluator: An End-to-End Guide from Standard Benchmarks to Custom Datasets

This guide shows how to use NVIDIA NeMo Evaluator in a PAASUP DIP environment. It covers the end-to-end process of evaluating LLMs connected via an NIM Proxy, using both standard benchmarks and custom data, from setup to result interpretation.

Deploying NVIDIA NIM Microservice in a Private Environment

This guide explains how to securely and efficiently deploy LLM models like llama-3.2-3b-instruct in a private environment using NVIDIA NIM Microservice on PAASUP DIP. It supports both Python code integration and no-code workflows with Flowise, ideal for diverse enterprise AI applications.