DOTNET: Understanding Garbage Collection with Gen0, Gen1, Gen2

Yes, we can absolutely do this in one self-contained program 👍
You’ll be able to:

• See GC before/after each scenario
• Trigger Gen0 / Gen1 / Gen2 collections
• Observe heap size and collection counts
• Use this with dotnet-counters / dotMemory / VS profiler if you want

Below is a single Program.cs you can drop into a new console app.

1️⃣ Create the project

dotnet new console -n GcGenerationsDemo
cd GcGenerationsDemo
Code language: JavaScript (javascript)

Replace all contents of Program.cs with this:

2️⃣ How to run and “experience” GC

From inside the project folder:

dotnet run

The program will walk you through:

1. Startup stats
2. Gen0 scenario – short-lived allocations
3. Gen1 scenario – medium-lived allocations
4. Gen2 scenario – long-lived allocations

Each step prints:

• Gen0/Gen1/Gen2 collection counts
• Total managed heap (MB)

You’ll see:

• Gen0 counts jump a lot in the first scenario.
• Gen1 counts increase in the second scenario.
• Gen2 counts and heap size behavior change in the third scenario.

3️⃣ Optional: Watch it live with dotnet-counters

In another terminal, while the app is running:

1. Get process list: dotnet-counters ps
2. Find your GcGenerationsDemo PID, then: dotnet-counters monitor --process-id <PID> System.Runtime

Watch:

• gc-heap-size
• gen-0-gc-count
• gen-1-gc-count
• gen-2-gc-count

Run each scenario and you’ll see the counters move in sync with console output.

4️⃣ How this maps to Gen0 / Gen1 / Gen2 concepts

• Gen0 scenario
  • Many tiny, short-lived objects
  • Mostly collected in Gen0
  • You’ll see Gen0 collections spike
• Gen1 scenario
  • Objects kept alive briefly in a List<byte[]>
  • They survive at least one collection → promoted to Gen1
  • We force Gen1 collections and then free references
  • You see Gen1 counts increase, heap shrink
• Gen2 scenario
  • Objects stored in a static list (LongLivedHolder.Buffers)
  • They are long-lived; promoted to Gen2
  • Even after Gen2 collection, many remain because references are still held
  • This is how leaks and long-lived caches behave

1️⃣ How to read the four numbers

Each PrintGcStats gives you:

• Gen0 collections – how many times .NET cleaned short-lived objects
• Gen1 collections – how many times it cleaned objects that survived Gen0
• Gen2 collections – how many times it cleaned long-lived objects
• Total managed heap – how much managed memory is still in use after GC (roughly)

These counters are cumulative since process start.

2️⃣ Gen0 scenario – short-lived objects

You saw:

=== Gen0 Scenario: Short-lived objects ===
--- GC Stats: Before Gen0 work ---
  Gen0 collections: 0
  Gen1 collections: 0
  Gen2 collections: 0
  Total managed heap: 0 MB

Gen0 scenario completed in 35 ms
--- GC Stats: After Gen0 forced collection ---
  Gen0 collections: 71
  Gen1 collections: 1
  Gen2 collections: 0
  Total managed heap: 0 MB

What happened?

• We allocated millions of tiny objects in a loop.
• They were not stored anywhere, so they died quickly.
• GC cleaned them mostly in Gen0:
  • Gen0 collections: 0 → 71 ✅
• A few objects survived briefly → promoted to Gen1 once:
  • Gen1 collections: 0 → 1
• After the forced collection:
  • Total managed heap: 0 MB (rounded) → means almost nothing left.

👉 Interpretation:

“Lots of short-lived garbage → GC handled it cheaply in Gen0.
We generated a ton of allocations, but the memory was fully reclaimed. No leak, GC working as designed.”

This is typical of request-scoped allocations in APIs when they’re well-behaved.

3️⃣ Gen1 scenario – medium-lived objects

You saw:

=== Gen1 Scenario: Medium-lived objects ===
--- GC Stats: Before Gen1 work ---
  Gen0 collections: 71
  Gen1 collections: 1
  Gen2 collections: 0
  Total managed heap: 0 MB

Gen1 scenario completed in 2613 ms
--- GC Stats: After Gen1 forced collections ---
  Gen0 collections: 627
  Gen1 collections: 296
  Gen2 collections: 33
  Total managed heap: 3131 MB

What did the code do here?

• It allocated a lot of 16 KB buffers and stored them in a List<byte[]> survivors.
• That local list stayed alive for a while → the buffers survived multiple Gen0 collections.
• That caused:
  • Gen0 collections: 71 → 627 (lots of allocations)
  • Gen1 collections: 1 → 296 (many promotions & cleanups)
  • Gen2 collections: 0 → 33 (some long-lived promotions too)

Why is heap so big here (3131 MB)?

• Right after the scenario, before the runtime has fully compacted and reused memory, GetTotalMemory sees ~3 GB still reserved/used.
• These objects were just cleared at the end of the scenario (we call survivors.Clear() and GC.Collect), but this snapshot is still showing that a lot of memory was in play.

Then before the next scenario you saw:

--- GC Stats: Before Gen2 work ---
  ...
  Total managed heap: 2 MB

So eventually the runtime fully reclaimed it, and the heap dropped back down.

👉 Interpretation:

“Here we created objects that lived longer than Gen0 (in a list).
We see big jumps in Gen1 and Gen2 collections and temporary heap growth (~3 GB).
After clearing references and more GC, memory drops back to a few MB → no leak, just heavy temporary pressure.”

This demonstrates:

• Promotion from Gen0 → Gen1 → Gen2
• Longer-lived objects = more expensive GC
• Why you don’t want to keep large collections alive longer than necessary.

4️⃣ Gen2 scenario – long-lived / leaked objects

You saw:

=== Gen2 Scenario: Long-lived objects ===
--- GC Stats: Before Gen2 work ---
  Gen0 collections: 627
  Gen1 collections: 296
  Gen2 collections: 33
  Total managed heap: 2 MB

Gen2 scenario completed in 184 ms
--- GC Stats: After Gen2 forced collection ---
  Gen0 collections: 693
  Gen1 collections: 329
  Gen2 collections: 34
  Total managed heap: 392 MB

Long-lived objects stored: 50000
Note: Because we still keep references, these objects won't be freed.

What did the code do here?

• Allocated 50,000 × 8 KB buffers ≈ ~400 MB.
• Stored them in LongLivedHolder.Buffers, which is static.
• We never clear that list → those objects are effectively long-lived.

Even after a full Gen2 collection:

• Gen2 collections: 33 → 34 (we forced a full GC)
• But heap only drops to:
  • Total managed heap: 392 MB
• And we still have:
  • Long-lived objects stored: 50000

👉 Interpretation:

“These are long-lived objects (or a leak).
Even after a full Gen2 GC, almost 400 MB remains because we’re still holding references in a static list.
This is what a memory leak / long-lived cache looks like in production:
Gen2 collections happen, but memory never really goes down.”

This is the pattern you’d see in:

• Static caches that don’t evict
• Static lists/dicts that only grow
• Singletons holding on to big data
• Event handler leaks, etc.

5️⃣ How to explain (short version)

“In the Gen0 scenario, we created millions of tiny, short-lived objects.
GC handled them mostly in Gen0 (71 collections), and after GC, memory is basically 0 MB.
This is healthy, short-lived garbage.

In the Gen1 scenario, we kept objects alive for a while in a list.
They survived Gen0, got promoted to Gen1 and some to Gen2.
We see big Gen1/Gen2 collection counts and temporary heap growth to ~3 GB, but after clearing references, the heap returns to a small size.
This shows the cost of medium-lived objects.

In the Gen2 scenario, we stored 50k buffers in a static list.
Even after a full Gen2 collection, we still use ~392 MB.
Because the app still references these objects, the GC cannot free them.
This is exactly how long-lived objects and memory leaks behave in real .NET apps.”

Rajesh Kumar

I’m a DevOps/SRE/DevSecOps/Cloud Expert passionate about sharing knowledge and experiences. I have worked at Cotocus. I share tech blog at DevOps School, travel stories at Holiday Landmark, stock market tips at Stocks Mantra, health and fitness guidance at My Medic Plus, product reviews at TrueReviewNow , and SEO strategies at Wizbrand.

Do you want to learn Quantum Computing?

Please find my social handles as below;

Rajesh Kumar Personal Website
Rajesh Kumar at YOUTUBE
Rajesh Kumar at INSTAGRAM
Rajesh Kumar at X
Rajesh Kumar at FACEBOOK
Rajesh Kumar at LINKEDIN
Rajesh Kumar at WIZBRAND

Rajesh Kumar DailyLogs