Sawmills AI, a San Francisco-based startup, has recently emerged from stealth mode to tackle the increasing costs associated with enterprise observability. By leveraging large language models (LLMs) and proprietary machine learning (ML) models, the company aims to reduce the data volume sent to observability vendors while ensuring organizations retain all their original data.
The Growing Importance of Observability
Evolution of Data Observability
The practice of data observability, which involves using software tools to monitor the functioning of an organization’s entire software suite, dates back to the late 1950s but has gained renewed importance in the generative AI era. Observability platform vendors such as Splunk and Datadog have built multibillion-dollar businesses by helping enterprises organize telemetry data, which indicates the status of different processes and whether everything is functioning correctly. However, many organizations now find themselves dealing with an overwhelming amount of data and the associated costs.
This evolution has been driven largely by the advancements in artificial intelligence, machine learning, and big data analytics, which have made it possible to collect, process, and analyze vast amounts of telemetry data in near real-time. The current landscape of observability is vastly different from its early days, with modern tools offering extensive capabilities to track everything from application performance to infrastructure health. Yet, with these advancements comes a sharp increase in the volume of data generated, which in turn has resulted in ballooning costs for enterprises trying to maintain observability.
Rising Costs and Industry Concerns
A 2023 survey revealed that 98% of companies experienced unexpected increases in observability expenses, with 51% facing overages on a monthly basis. Charity Majors, co-founder and CTO of Honeycomb, suggests that organizations should allocate 20 to 30% of their infrastructure budget to observability, highlighting the significant financial impact on businesses.
The rising costs are not just a financial burden but also a strategic concern for enterprise leadership. Unpredictable expenses make it difficult for companies to budget and plan for their technology investments. Furthermore, the exponential growth in telemetry data has made it increasingly challenging for organizations to manage and extract meaningful insights efficiently. Consequently, IT departments are overwhelmed by data overload, leading to delays in identifying and resolving issues, and ultimately affecting overall system reliability and performance. This pressing need for a more efficient approach to managing observability data has set the stage for innovative solutions like Sawmills AI.
Sawmills AI’s Approach to Cost Reduction
Middleware Layer for Optimization
Sawmills AI acts as a middleware layer between observability platforms and their customers, using AI and ML to filter, consolidate, and optimize telemetry data before it is sent to tools like Datadog, Splunk, or New Relic. This approach helps reduce data volume and associated costs significantly. By intelligently processing data at this intermediary stage, Sawmills AI can deliver streamlined and relevant telemetry information, ensuring that only the most pertinent data reaches the observability tools.
This layered approach also means that Sawmills AI can retrofit into existing observability workflows without requiring complete overhauls of the systems in place. Organizations can continue to use their preferred observability tools while gaining the advantage of reduced data volumes and cost savings. Additionally, the ability to filter and optimize data before sending it to observability vendors can ease the burden on internal teams, allowing them to focus on analyzing actionable insights rather than drowning in an excess of raw telemetry data.
Funding and Market Validation
Recently, Sawmills AI raised $10 million in seed funding led by Team8, with participation from Mayfield and Alumni Ventures. The startup’s co-founders, including CTO Amir Jakoby and CPO Erez Rusovsky, initially set out to solve a different problem but shifted focus after conversations with industry stakeholders highlighted the rising cost of observability solutions. The strong support from investors underscores the market’s recognition of the urgent need for cost-effective observability solutions.
The successful funding round not only validates the effectiveness of Sawmills AI’s approach but also provides the necessary resources for further innovation and market expansion. With this injection of capital, Sawmills AI plans to enhance their platform’s capabilities, hire top talent, and ramp up marketing efforts to reach a broader audience. This strategic investment demonstrates confidence in the startup’s potential to address a critical pain point in the industry and signals a promising future for the company as it continues to develop and refine its solutions.
Addressing Data Management Challenges
Centralized Telemetry Management
One major challenge identified was the lack of centralized telemetry management within organizations. Each developer typically generates their own telemetry data, leading to a proliferation of logs, metrics, and traces that are not managed efficiently. This fragmented approach contributes to data volumes growing unchecked, driving up costs and complicating troubleshooting efforts.
Without a centralized system to manage telemetry data, organizations struggle to maintain a coherent overview of their system performance. This disjointed data collection results in inefficiencies and redundancies, making it difficult to pinpoint issues accurately and quickly. Sawmills AI addresses these challenges by offering a comprehensive solution that brings all telemetry data under a unified management platform. This centralized approach not only streamlines data handling but also enables better analysis and more effective monitoring of system health, ultimately improving operational efficiency and reducing overall observability costs.
Smart Telemetry Data Management Platform
Sawmills AI offers a smart telemetry data management platform with several key features. These include a telemetry data explorer for full visibility into data flows, cost and availability control to understand the impact of telemetry data on expenses, and log and metric optimization to eliminate wasteful data processing. The platform also provides one-click actions for engineers to implement AI-driven recommendations instantly and vendor flexibility through support for OpenTelemetry, allowing seamless switching between observability vendors. Moreover, automated policy management ensures data governance, prevents overages, and maintains security compliance.
By using these tools, Sawmills AI claims to provide significant data volume reduction and associated cost savings for enterprise customers. For instance, converting millions of lines of logs into a single metric can reduce data volume by 10 to 100 times, significantly lowering costs. Additionally, the platform leverages leading LLMs, both open-source and from proprietary cloud vendors, to summarize and consolidate customer data before it is sent to their observability platforms.
Benefits and Early Adoption
Significant Data Volume Reduction
By converting millions of lines of logs into a single metric, Sawmills AI claims to reduce data volume by 10 to 100 times, significantly lowering costs. This remarkable reduction enables enterprises to transmit essential data to observability platforms without the burden of unnecessary information. The ability to distill vast amounts of telemetry data into concise, actionable metrics is a game-changer for organizations managing complex IT environments.
Sawmills AI’s data reduction capabilities also translate directly into more predictable and manageable costs. Organizations no longer have to grapple with unexpected spikes in observability expenses, as the platform ensures only the most critical data is forwarded on to observability tools. This streamlined approach not only reduces costs but also enhances the overall efficiency and effectiveness of system monitoring, as engineers can focus on interpreting and acting upon meaningful insights rather than sifting through extraneous data.
Enhanced Data Quality and Governance
Early adopters like Edi Buslovich, VP of engineering at Via, have experienced benefits such as optimized telemetry data, reduced costs, and improved governance. Sawmills AI enhances the value derived from existing tools by allowing enterprises to control their telemetry data more efficiently. With enhanced data quality, companies can achieve better system performance, faster issue resolution, and stronger data governance practices.
Improved data governance is particularly crucial in today’s regulatory environment, where compliance with data protection and privacy laws is paramount. Sawmills AI’s automated policy management features help organizations maintain compliance effortlessly, ensuring that data handling practices meet stringent security and governance standards. Early successes with the platform have demonstrated its ability to deliver tangible benefits, reinforcing Sawmills AI’s position as a valuable ally for enterprises seeking to optimize their observability strategies and achieve cost savings without sacrificing performance or compliance.
Future Prospects and Market Expansion
Target Market and Customer Base
Sawmills AI targets mid-to-large enterprises with 500 to 5,000 employees, particularly those with significant investments in cloud and observability. The company has already secured dozens of paying customers and plans to expand its market reach following its public launch. By focusing on organizations with substantial telemetry data and observability needs, Sawmills AI positions itself as an essential solution for businesses looking to streamline their operations and reduce costs.
The startup’s tailored approach allows it to address the unique challenges faced by larger enterprises, such as managing extensive telemetry data and balancing cost-efficiency with performance. As Sawmills AI continues to expand its presence in the market, the company aims to build on its early successes by enhancing its platform’s capabilities, addressing customer feedback, and forging strategic partnerships with leading observability vendors. This forward-thinking strategy ensures that Sawmills AI remains at the forefront of the observability landscape, providing innovative solutions that meet the evolving needs of its growing customer base.
Continuous Improvement and Demand
Sawmills AI, a San Francisco startup, has recently come out of stealth mode with a mission to address the rising expenses linked to enterprise observability. By harnessing the power of large language models (LLMs) and their own proprietary machine learning (ML) models, Sawmills AI strives to lower the amount of data sent to observability vendors while allowing organizations to keep their complete original data intact.
Ronit Belson, co-founder and CEO, points out that Sawmills AI is dedicated to granting customers control over their telemetry data, deviating from the traditional dependence on observability vendors. This approach not only reduces costs but also enhances data management and security for businesses.
The emergence of Sawmills AI marks a significant step in the enterprise observability landscape, illustrating a shift towards more efficient and cost-effective solutions. By focusing on customer empowerment and data retention, the startup is poised to make a substantial impact in the field.
Belson’s vision for Sawmills AI not only addresses cost concerns but also pushes forward innovation in data management within enterprises. This aligns with the broader trend of leveraging advanced AI and ML technologies to solve complex business challenges, offering tailored solutions that meet the specific needs of modern organizations.