Bridging Operational Technology (OT) and Enterprise Analytics: A Framework for Integrating AVEVA PI with Cloud-Scale ELT Pipelines
Open accessAbstract
Industrial organizations struggle to bridge the gap between high frequency operational telemetry and cloud-scale enterprise analytics without sacrificing semantic integrity and incurring prohibitive computational costs. While the AVEVA PI System remains a mission-critical staple for contextualizing time-series data, traditional integration methods often trigger significant semantic loss during the transition to the cloud. This research introduces a reference framework utilizing Informatica IDMC and a layered Extract, Load, Transform (ELT) architecture to synchronize the PI System with Snowflake. Unlike standard migrations that flatten data into disconnected tags, this approach embeds operational semantics directly through the preservation of asset models and event context. By utilizing warehouse native Change Data Capture (CDC), specifically Snowflake Streams and Tasks, the framework replaces inefficient full refresh cycles with high performance incremental processing. To address the rigorous requirements of critical infrastructure, the architecture aligns with NIST SP 800-82 Revision 3 security standards and introduces a quantitative observability model based on Service Level Objectives (SLOs) for data freshness and reconciliation. I validated the framework through a utility. y scale implementation, which yielded a 95th percentile latency of 7.4 minutes and approximately 38% reduction in compute consumption compared to legacy methods. Ultimately, the aim of this research is to provide a practical, reproducible blueprint for organizations seeking to modernize industrial analytics while maintaining operational trust, performance, and continuity.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
IOPF extends SLA monitoring to proximity-based alerting and cascade risk quantification across pipeline dependency graphs.
“IOPF extends SLA monitoring to proximity-based alerting and cascade risk quantification across pipeline dependency graphs.”
From Data Observability Framework for Mission-Critical Industrial ETL Pipelines: Anomaly Detection, SLA Modeling, and Proactive Failure Identification (2026) · cites this paper as [5]
Section 7.3 (page unavailable) describes an automated SLA heatmap for monitoring pipeline health and identifying failure hotspots. The paper does not describe IOPF, proximity-based alerting, or cascade-risk quantification across dependency graphs; those extensions are unsupported by our copy. (LLM, conf=0.84)
Experimental — accuracy not yet measuredThe MTTR figures reported measure observability-driven incident resolution across the full IOPF deployment and differ from the integration-pipeline MTTR figures reported in [5].
“The MTTR figures reported here measure observability-driven incident resolution across the full IOPF deployment. They differ from the integration-pipeline MTTR figures reported in [5], which were collected over a 30-day integration validation window using SLO-based alerting only.”
From Data Observability Framework for Mission-Critical Industrial ETL Pipelines: Anomaly Detection, SLA Modeling, and Proactive Failure Identification (2026) · cites this paper as [5]
Section 8.2 reports a mean MTTR of 22 minutes from a 30-day evaluation against the framework’s SLOs, and Section 7.2 defines MTTR as time to recover from failure. The source does not say that the figures were collected using SLO-based alerting only. The comparison with the citing paper’s IOPF deployment is the citing authors’ own framing, not a finding attributed to this source. Page numbers are unavailable in our copy. (LLM, conf=0.84)
Experimental — accuracy not yet measuredThe contribution of IOPF is a framework that makes the operational behavior of industrial data pipelines auditable in the same way the integrated platform is itself auditable under [5].
“The contribution of IOPF is a framework that makes the operational behavior of those pipelines auditable in the same way the integrated platform is itself auditable under [5].”
From Data Observability Framework for Mission-Critical Industrial ETL Pipelines: Anomaly Detection, SLA Modeling, and Proactive Failure Identification (2026) · cites this paper as [5]
IOPF complements the NIST SP 800-82-aligned security controls established in [5].
“IOPF complements the NIST SP 800-82-aligned security controls established in [5].”
From Data Observability Framework for Mission-Critical Industrial ETL Pipelines: Anomaly Detection, SLA Modeling, and Proactive Failure Identification (2026) · cites this paper as [5]
The four-class volume anomaly model extends the binary reconciliation rate metric (parity met / parity broken) from the prior framework into a four-tier classification with distinct alert priorities and response protocols.
“This four-class model extends the binary reconciliation rate metric (parity met / parity broken) from the prior framework into a four-tier classification with distinct alert priorities and response protocols.”
From Data Observability Framework for Mission-Critical Industrial ETL Pipelines: Anomaly Detection, SLA Modeling, and Proactive Failure Identification (2026) · cites this paper as [5]
The prior framework [5] established security alignment for the integration layer and addressed how to bridge OT and IT data worlds at integration time.
“The prior framework [5] established security alignment for the integration layer; IOPF extends the underlying telemetry model so that audit records satisfy operational diagnostic needs as well as documentation requirements that critical infrastructure environments routinely face [19].”
From Data Observability Framework for Mission-Critical Industrial ETL Pipelines: Anomaly Detection, SLA Modeling, and Proactive Failure Identification (2026) · cites this paper as [5]
The t_lat = T_pub - T_src formulation from prior work is generalized into a per-pipeline freshness threshold λ(p_i) calibrated against historical lag distribution.
“The t lat = T pub - T src formulation from [5], where T pub is the target availability timestamp and T src is the source timestamp is generalized here into a per-pipeline freshness threshold λ(p ᵢ ) calibrated against historical lag distribution.”
From Data Observability Framework for Mission-Critical Industrial ETL Pipelines: Anomaly Detection, SLA Modeling, and Proactive Failure Identification (2026) · cites this paper as [5]
The principle of categorizing failures rather than treating them as binary success/fail came from prior work [5].
“The principle of categorizing failures rather than treating them as binary success/fail came from the prior work [5]”
From Data Observability Framework for Mission-Critical Industrial ETL Pipelines: Anomaly Detection, SLA Modeling, and Proactive Failure Identification (2026) · cites this paper as [5]
The new contribution of this paper is the explicit class-to-protocol mapping that allows a junior engineer to follow the same diagnostic path as the author.
“the new contribution here is the explicit class-to-protocol mapping that allows a junior engineer to follow the same diagnostic path I would.”
From Data Observability Framework for Mission-Critical Industrial ETL Pipelines: Anomaly Detection, SLA Modeling, and Proactive Failure Identification (2026) · cites this paper as [5]
The Introduction and Section 10 describe a cloud-scale industrial telemetry framework and name its contributions as semantic continuity, architectural efficiency, and standardized reliability. Searches for class-to-protocol mapping and a junior engineer’s diagnostic path found no passage addressing either point in our copy. The different list of contributions alone does not establish a contradiction. (LLM, conf=0.75)
Experimental — accuracy not yet measured
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Cites
No references could be resolved yet.
Research works printed but not linked: 2
- S. Karumuri, F. Solleza, S. Zdonik, and N. Tatbul, "Towards Observability Data Management at Scale," SIGMOD Record, 2020. [Online]. Available: https://people.csail.mit.edu/tatbul/publications/sigmod_r ecord20.pdf.
- Reddy, R. R. P. (2024). Enhancing endpoint security through collaborative zero-trust integration: a multi- agent approach. International Journal of Computer Trends and Technology, 72(8), 86-90.
Cited by
Sorted by citations
Sojitra, E. (2026) · American International Journal of Computer Science and Technology · doi:10.63282/3117-5481/aijcst-v8i3p101
Self-citation
- Authors
- Sojitra, E.
- Corresponding author
- Sojitra, Ekta
- Venue
- International Journal of AI, BigData, Computational and Management Studies
- Publisher
- ScienceTech Xplore
- Published
- 2026
- Volume
- 7
- Pages
- 244-250
- ISSN
- 3050-9416 (ISSN-L)
- OpenAlex
- W7152964052
- Cited by
- 0
- References
- 0
- FWCI
- 0.00
- Type
- journal article
- Language
- en
- Open access
- bronze
- doi.org
- Locations
- International Journal of AI BigData Computational and Management Studies (publishedVersion)
- Topics
- Physical Sciences › Engineering › Industrial and Manufacturing Engineering › Digital Transformation in Industry
- Social Sciences › Business, Management and Accounting › Management Information Systems › Big Data and Business Intelligence
- Physical Sciences › Computer Science › Computer Networks and Communications › Advanced Database Systems and Queries
- Keywords
- cloud data integration, snowflakes, change data capture, incremental data processing, semantic integrity, critical infrastructure security, service level objectives, data freshness, industrial analytics
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
