Picdmo
Picdmo Revolutionizes Photo Management with Zilliz Cloud
![Picdmo](https://assets.zilliz.com/picdmo_detail_page_ae60e0c4a7.png)
90% reduction
in search latency
Sharp reduction
on $ and time
Easily scale
as business grows
When we conducted our first test run (on Zilliz Cloud), the performance improvement was astonishing. Our search time went from eight seconds down to sub one second. We were nearly falling off our chairs with amazement at the speed.
Alex Alexander
About Picdmo
Picdmo, founded in 2020 by Alex Alexander and Gerry Speranza, is on a mission to revolutionize photo management. Recognizing the universal struggle of finding and organizing digital photos, the duo set out to create an AI-powered solution. The catalyst for Picdmo's inception was a simple yet frustrating experience when Alex couldn't locate a cherished image from two and a half years ago. This common dilemma, faced by individuals of all ages and across various devices, inspired them to develop an "everything app" available on Android and iOS.
Picdmo's vision is to provide a seamless photo organization and search experience, addressing the widespread problem of image chaos. The heart of Picdmo lies in its advanced AI engine, painstakingly developed over three and a half years and patented for its effectiveness. This case study explores how Picdmo leverages Zilliz Cloud's vector search technology to enhance its AI-powered platform and shape the future of photo management.
Challenges: Navigate the Path to Efficient Search Performance
Building an innovative photo management application like Picdmo presented a journey filled with technological hurdles.
One of the significant challenges arose when Picdmo was working on implementing a natural language search feature and refining search results. The performance of their existing database proved to be subpar for the specific demands of their AI-powered platform, particularly when dealing with embeddings generated by the AI engine.
The issue became evident as search queries took an unacceptable 8 to 10 seconds to return results, a timeframe far from the desired subsecond response expected by users. In response to this challenge, Picdmo initially explored the path of setting up their own servers with Milvus, a popular open-source vector database. While they made some progress in reducing the response time to four seconds, the endeavor proved labor-intensive and financially costly. To make matters more perplexing, the response time unexpectedly jumped back to eight seconds, necessitating a week-long effort to diagnose and rectify configuration issues.
Solutions: Elevated Search Performance with Zilliz Cloud
When Picdmo encountered the challenge of sluggish search performance, their journey led them to explore Zilliz Cloud, the fully managed Milvus service, as a potential solution.
Their primary goal was to enhance search performance, which initially yielded results in an excruciating eight seconds. The breakthrough occurred when Picdmo conducted a test run using Zilliz Cloud. The results were astounding, with response times reduced from eight seconds to less than one second, even with extreme data loads. This exceptional performance improvement left no doubt that Zilliz Cloud was the solution to their search performance challenges. More importantly, Zilliz Cloud ensures swift search performance and provides the scalability needed to maintain efficiency even under heavy loads, aligning perfectly with Picdmo's goals for a streamlined and responsive photo management experience.
Picdmo collaborated closely with Zilliz's technical experts to implement a more efficient system throughout this technological endeavor. Zilliz Cloud assumed a central role in their infrastructure, enabling Picdmo to deliver the required search performance. The Picdmo smartphone app uploads images to the Picdmo cloud, where the AI engine analyzes them, generating embeddings.
These embeddings are stored in Zilliz Cloud, forming the cornerstone of Picdmo's technology representation of media. When users initiate searches, Picdmo's AI engine expertly transforms search phrases into embeddings, efficiently processed in Zilliz Cloud's vector database. This architecture consistently delivers subsecond responses to users' queries.
Results: Enhanced Search, Reduced Operational Burden, and Better User experience
Adopting Zilliz Cloud was a transformative decision for Picdmo, with far-reaching business impact. Beyond the technical enhancements, the results were evident in various aspects of Picdmo's operations.
First and foremost, search performance dramatically improved. What was once an eight-second wait for search results became a subsecond response time, exceeding user expectations and significantly enhancing the user experience. Picdmo's users could now find and organize their photos with unprecedented speed and efficiency.
From a business perspective, Picdmo experienced substantial time and cost savings by integrating Zilliz Cloud into its architecture. It eliminated the need to set up their vector databases, streamlining their technological infrastructure and reducing operational complexities.
Furthermore, Zilliz's support played a crucial role in addressing challenges due to Picdmo's rapid growth. Their responsiveness and problem-solving capabilities ensured minimal downtime and a seamless user experience, contributing to Picdmo's reputation for reliability.
By resolving Picdmo's technical challenges and delivering tangible business results, Zilliz has become an indispensable partner in Picdmo's journey to revolutionize photo management, which enhances user satisfaction, reduces costs, and ensures scalability for future growth.
Future Plans: Expanding with Zilliz Collaboration
Picdmo's journey to becoming a comprehensive multimedia application includes introducing diverse new features. This expansion encompasses photos, videos, and a range of other capabilities.
As these enhancements roll out, they'll place evolving demands on Zilliz Cloud, Picdmo's trusted technology partner. Picdmo will hold specialized technology meetings with Zilliz's team to ensure smooth product development and peak performance. These sessions will discuss upcoming features and how Zilliz Cloud could support them.
This proactive approach ensures that Picdmo and Zilliz can address potential performance challenges and optimize the platform, maintaining the high performance and reliability users expect as Picdmo's feature set grows.