5 Takeaways from OTEL Day in Seattle: A Glimpse into the Evolution of Observability

5 Takeaways from OTEL Day in Seattle: A Glimpse into the Evolution of Observability

5 Takeaways from OTEL Day in Seattle: A Glimpse into the Evolution of Observability

High Volume Context Aware APM

End-to-end Distributed Tracing and Service-centric Observability at Scale

Highly context-aware APM powered by distributed tracing with built-in RED metrics, richly augmented service map visualizations, and request flows with high retention of traces and metrics derived from them.

No Sampling Required

Collect, and analyze distributed traces with

a long chain of spans in an unsampled way

and store them for longer retention. All

detailed attributes of the chains of spans

for all traces are stored and indexed.

Ingest traces in Zipkin, Jaeger,

OpenTelemetry, OpenTracing, Datadog

and Elastic agent formats, but store in

normalized way.

Any Stack Anywhere

APM on hosts, containers, serverless

functions using 100+ integrations with

third party frameworks, libraries in various

languages Java, .NET, PHP, Node.js, Ruby,

Python, Go, or C++ applications.

Built-In RED metrics

Kloudfuse APM builds a detailed

breakdown of requests, error, latency,

throughput, and saturation metrics across

all the services, methods, endpoints, edges,

and resources along the spans in the

traces.

Rich Visualizations

Rich and context-aware service map

visualizations, request flow maps, and

flame-graphs on the fine-grained spans of

past and live traces augmented with the

RED metrics provide the highest context

troubleshooting experience.

100x cost reduction

Take advantage of the same built-in

storage efficiency and ingestion

optimization, to store and query traces and

metrics associated with them.

Cost Analysis

# of traces/minute

1,000,000

10,000,000

100,000,000

1,000,000,000

Compute Cost

$1,042

$7,289

$69,246

$691,939

Storage Cost

$119

$988

$9,644

$96,410

Total Cost*

$1,161

$8,277

$78,890

$788,349

* Cost is per month and based on GCP node pricing and high ingestion speed for spans per core, with minimal bytes per span and

dedicated cores for high performant queries and cores for replication of data with 7 days of retention with no Sampling

Cost Analysis

# of traces/minute

1,000,000

10,000,000

100,000,000

1,000,000,000

Compute Cost

$1,042

$7,289

$69,246

$691,939

Storage Cost

$119

$988

$9,644

$96,410

Total Cost*

$1,161

$8,277

$78,890

$788,349

* Cost is per month and based on GCP node pricing and high ingestion speed for spans per core, with minimal bytes per span and

dedicated cores for high performant queries and cores for replication of data with 7 days of retention with no Sampling

No Sampling Required

Collect, and analyze distributed traces with

a long chain of spans in an unsampled way

and store them for longer retention. All

detailed attributes of the chains of spans

for all traces are stored and indexed.

Ingest traces in Zipkin, Jaeger,

OpenTelemetry, OpenTracing, Datadog

and Elastic agent formats, but store in

normalized way.

Any Stack Anywhere

APM on hosts, containers, serverless

functions using 100+ integrations with

third party frameworks, libraries in various

languages Java, .NET, PHP, Node.js, Ruby,

Python, Go, or C++ applications.

Built-In RED metrics

Kloudfuse APM builds a detailed

breakdown of requests, error, latency,

throughput, and saturation metrics across

all the services, methods, endpoints, edges,

and resources along the spans in the

traces.

Rich Visualizations

Rich and context-aware service map

visualizations, request flow maps, and

flame-graphs on the fine-grained spans of

past and live traces augmented with the

RED metrics provide the highest context

troubleshooting experience.

100x cost reduction

Take advantage of the same built-in

storage efficiency and ingestion

optimization, to store and query traces and

metrics associated with them.

No Sampling Required

Collect, and analyze distributed traces with

a long chain of spans in an unsampled way

and store them for longer retention. All

detailed attributes of the chains of spans

for all traces are stored and indexed.

Ingest traces in Zipkin, Jaeger,

OpenTelemetry, OpenTracing, Datadog

and Elastic agent formats, but store in

normalized way.

Any Stack Anywhere

APM on hosts, containers, serverless

functions using 100+ integrations with

third party frameworks, libraries in various

languages Java, .NET, PHP, Node.js, Ruby,

Python, Go, or C++ applications.

Built-In RED metrics

Kloudfuse APM builds a detailed

breakdown of requests, error, latency,

throughput, and saturation metrics across

all the services, methods, endpoints, edges,

and resources along the spans in the

traces.

Rich Visualizations

Rich and context-aware service map

visualizations, request flow maps, and

flame-graphs on the fine-grained spans of

past and live traces augmented with the

RED metrics provide the highest context

troubleshooting experience.

100x cost reduction

Take advantage of the same built-in

storage efficiency and ingestion

optimization, to store and query traces and

metrics associated with them.

Cost Analysis

# of traces/minute

1,000,000

10,000,000

100,000,000

1,000,000,000

Compute Cost

$1,042

$7,289

$69,246

$691,939

Storage Cost

$119

$988

$9,644

$96,410

Total Cost*

$1,161

$8,277

$78,890

$788,349

* Cost is per month and based on GCP node pricing and high ingestion speed for spans per core, with minimal bytes per span and

dedicated cores for high performant queries and cores for replication of data with 7 days of retention with no Sampling

Cost Analysis

# of traces/minute

1,000,000

10,000,000

100,000,000

1,000,000,000

Compute Cost

$1,042

$7,289

$69,246

$691,939

Storage Cost

$119

$988

$9,644

$96,410

Total Cost*

$1,161

$8,277

$78,890

$788,349

* Cost is per month and based on GCP node pricing and high ingestion speed for spans per core, with minimal bytes per span and

dedicated cores for high performant queries and cores for replication of data with 7 days of retention with no Sampling

Observe. Analyze. Automate

Observe. Analyze. Automate

Observe. Analyze. Automate