diff --git a/app/vmagent/main.go b/app/vmagent/main.go
index 7eedc3a24..e7c9b92a8 100644
--- a/app/vmagent/main.go
+++ b/app/vmagent/main.go
@@ -8,7 +8,6 @@ import (
 	"net/http"
 	"os"
 	"strings"
-	"sync/atomic"
 	"time"
 
 	"github.com/VictoriaMetrics/VictoriaMetrics/app/vmagent/csvimport"
@@ -459,7 +458,7 @@ func requestHandler(w http.ResponseWriter, r *http.Request) bool {
 		w.WriteHeader(http.StatusOK)
 		return true
 	case "/ready":
-		if rdy := atomic.LoadInt32(&promscrape.PendingScrapeConfigs); rdy > 0 {
+		if rdy := promscrape.PendingScrapeConfigs.Load(); rdy > 0 {
 			errMsg := fmt.Sprintf("waiting for scrapes to init, left: %d", rdy)
 			http.Error(w, errMsg, http.StatusTooEarly)
 		} else {
diff --git a/app/vminsert/main.go b/app/vminsert/main.go
index 78fee4ead..a5d54553c 100644
--- a/app/vminsert/main.go
+++ b/app/vminsert/main.go
@@ -6,7 +6,6 @@ import (
 	"fmt"
 	"net/http"
 	"strings"
-	"sync/atomic"
 	"time"
 
 	"github.com/VictoriaMetrics/metrics"
@@ -354,7 +353,7 @@ func RequestHandler(w http.ResponseWriter, r *http.Request) bool {
 		w.WriteHeader(http.StatusOK)
 		return true
 	case "/ready":
-		if rdy := atomic.LoadInt32(&promscrape.PendingScrapeConfigs); rdy > 0 {
+		if rdy := promscrape.PendingScrapeConfigs.Load(); rdy > 0 {
 			errMsg := fmt.Sprintf("waiting for scrape config to init targets, configs left: %d", rdy)
 			http.Error(w, errMsg, http.StatusTooEarly)
 		} else {
diff --git a/go.mod b/go.mod
index 01a23707b..d021148c0 100644
--- a/go.mod
+++ b/go.mod
@@ -8,7 +8,7 @@ require (
 	github.com/Azure/azure-sdk-for-go/sdk/storage/azblob v1.3.0
 	github.com/VictoriaMetrics/easyproto v0.1.4
 	github.com/VictoriaMetrics/fastcache v1.12.2
-	github.com/VictoriaMetrics/metrics v1.32.0
+	github.com/VictoriaMetrics/metrics v1.33.0
 	github.com/VictoriaMetrics/metricsql v0.74.0
 	github.com/aws/aws-sdk-go-v2 v1.25.0
 	github.com/aws/aws-sdk-go-v2/config v1.27.0
diff --git a/go.sum b/go.sum
index d4874d507..8e7976aa6 100644
--- a/go.sum
+++ b/go.sum
@@ -69,8 +69,8 @@ github.com/VictoriaMetrics/easyproto v0.1.4/go.mod h1:QlGlzaJnDfFd8Lk6Ci/fuLxfTo
 github.com/VictoriaMetrics/fastcache v1.12.2 h1:N0y9ASrJ0F6h0QaC3o6uJb3NIZ9VKLjCM7NQbSmF7WI=
 github.com/VictoriaMetrics/fastcache v1.12.2/go.mod h1:AmC+Nzz1+3G2eCPapF6UcsnkThDcMsQicp4xDukwJYI=
 github.com/VictoriaMetrics/metrics v1.24.0/go.mod h1:eFT25kvsTidQFHb6U0oa0rTrDRdz4xTYjpL8+UPohys=
-github.com/VictoriaMetrics/metrics v1.32.0 h1:r9JK2zndYv0TIxFXLEHwhQqRdnu8/O3cwJiCBX4vJCM=
-github.com/VictoriaMetrics/metrics v1.32.0/go.mod h1:r7hveu6xMdUACXvB8TYdAj8WEsKzWB0EkpJN+RDtOf8=
+github.com/VictoriaMetrics/metrics v1.33.0 h1:EnkDEaGiL2u95t+W76GfecC/LMYpy+tFrexYzBWQIAc=
+github.com/VictoriaMetrics/metrics v1.33.0/go.mod h1:r7hveu6xMdUACXvB8TYdAj8WEsKzWB0EkpJN+RDtOf8=
 github.com/VictoriaMetrics/metricsql v0.74.0 h1:bVO7USXBBYEuEHQ3PZg/6216j0DvblZM+Q8sTRECkv0=
 github.com/VictoriaMetrics/metricsql v0.74.0/go.mod h1:k4UaP/+CjuZslIjd+kCigNG9TQmUqh5v0TP/nMEy90I=
 github.com/VividCortex/ewma v1.2.0 h1:f58SaIzcDXrSy3kWaHNvuJgJ3Nmz59Zji6XoJR/q1ow=
diff --git a/lib/backup/common/part.go b/lib/backup/common/part.go
index 8b9c68e31..84e3d1d70 100644
--- a/lib/backup/common/part.go
+++ b/lib/backup/common/part.go
@@ -46,14 +46,14 @@ func (p *Part) key() string {
 		// so it must have an unique key in order to always copy it during
 		// backup, restore and server-side copy.
 		// See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/5005
-		id := atomic.AddUint64(&uniqueKeyID, 1)
+		id := uniqueKeyID.Add(1)
 		return fmt.Sprintf("unique-%016X", id)
 	}
 	// Do not use p.FileSize in the key, since it cannot be properly initialized when resuming the restore for partially restored file
 	return fmt.Sprintf("%s%016X%016X%016X", p.Path, p.Offset, p.Size, p.ActualSize)
 }
 
-var uniqueKeyID uint64
+var uniqueKeyID atomic.Uint64
 
 // String returns human-readable representation of the part.
 func (p *Part) String() string {
diff --git a/lib/blockcache/blockcache.go b/lib/blockcache/blockcache.go
index 9192ffd70..c331f4227 100644
--- a/lib/blockcache/blockcache.go
+++ b/lib/blockcache/blockcache.go
@@ -172,14 +172,11 @@ func (c *Cache) cleanPerKeyMisses() {
 }
 
 type cache struct {
-	// Atomically updated fields must go first in the struct, so they are properly
-	// aligned to 8 bytes on 32-bit architectures.
-	// See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/212
-	requests uint64
-	misses   uint64
+	requests atomic.Uint64
+	misses   atomic.Uint64
 
 	// sizeBytes contains an approximate size for all the blocks stored in the cache.
-	sizeBytes int64
+	sizeBytes atomic.Int64
 
 	// getMaxSizeBytes() is a callback, which returns the maximum allowed cache size in bytes.
 	getMaxSizeBytes func() int
@@ -256,7 +253,7 @@ func (c *cache) RemoveBlocksForPart(p interface{}) {
 }
 
 func (c *cache) updateSizeBytes(n int) {
-	atomic.AddInt64(&c.sizeBytes, int64(n))
+	c.sizeBytes.Add(int64(n))
 }
 
 func (c *cache) cleanPerKeyMisses() {
@@ -281,7 +278,7 @@ func (c *cache) cleanByTimeout() {
 }
 
 func (c *cache) GetBlock(k Key) Block {
-	atomic.AddUint64(&c.requests, 1)
+	c.requests.Add(1)
 	var e *cacheEntry
 	c.mu.Lock()
 	defer c.mu.Unlock()
@@ -301,7 +298,7 @@ func (c *cache) GetBlock(k Key) Block {
 	}
 	// Slow path - the entry is missing in the cache.
 	c.perKeyMisses[k]++
-	atomic.AddUint64(&c.misses, 1)
+	c.misses.Add(1)
 	return nil
 }
 
@@ -367,7 +364,7 @@ func (c *cache) Len() int {
 }
 
 func (c *cache) SizeBytes() int {
-	return int(atomic.LoadInt64(&c.sizeBytes))
+	return int(c.sizeBytes.Load())
 }
 
 func (c *cache) SizeMaxBytes() int {
@@ -375,11 +372,11 @@ func (c *cache) SizeMaxBytes() int {
 }
 
 func (c *cache) Requests() uint64 {
-	return atomic.LoadUint64(&c.requests)
+	return c.requests.Load()
 }
 
 func (c *cache) Misses() uint64 {
-	return atomic.LoadUint64(&c.misses)
+	return c.misses.Load()
 }
 
 // lastAccessHeap implements heap.Interface
diff --git a/lib/blockcache/blockcache_timing_test.go b/lib/blockcache/blockcache_timing_test.go
index 776a29585..0fe8735ae 100644
--- a/lib/blockcache/blockcache_timing_test.go
+++ b/lib/blockcache/blockcache_timing_test.go
@@ -16,11 +16,11 @@ func BenchmarkKeyHashUint64(b *testing.B) {
 			h := k.hashUint64()
 			hSum += h
 		}
-		atomic.AddUint64(&BenchSink, hSum)
+		BenchSink.Add(hSum)
 	})
 }
 
-var BenchSink uint64
+var BenchSink atomic.Uint64
 
 func BenchmarkCacheGet(b *testing.B) {
 	c := NewCache(func() int {
diff --git a/lib/bloomfilter/limiter.go b/lib/bloomfilter/limiter.go
index 70069ddd8..117846c34 100644
--- a/lib/bloomfilter/limiter.go
+++ b/lib/bloomfilter/limiter.go
@@ -56,7 +56,7 @@ func (l *Limiter) MaxItems() int {
 // CurrentItems return the current number of items registered in l.
 func (l *Limiter) CurrentItems() int {
 	lm := l.v.Load()
-	n := atomic.LoadUint64(&lm.currentItems)
+	n := lm.currentItems.Load()
 	return int(n)
 }
 
@@ -72,7 +72,7 @@ func (l *Limiter) Add(h uint64) bool {
 }
 
 type limiter struct {
-	currentItems uint64
+	currentItems atomic.Uint64
 	f            *filter
 }
 
@@ -83,12 +83,12 @@ func newLimiter(maxItems int) *limiter {
 }
 
 func (l *limiter) Add(h uint64) bool {
-	currentItems := atomic.LoadUint64(&l.currentItems)
+	currentItems := l.currentItems.Load()
 	if currentItems >= uint64(l.f.maxItems) {
 		return l.f.Has(h)
 	}
 	if l.f.Add(h) {
-		atomic.AddUint64(&l.currentItems, 1)
+		l.currentItems.Add(1)
 	}
 	return true
 }
diff --git a/lib/bytesutil/bytesutil_timing_test.go b/lib/bytesutil/bytesutil_timing_test.go
index c140a50fe..6ba8b0d90 100644
--- a/lib/bytesutil/bytesutil_timing_test.go
+++ b/lib/bytesutil/bytesutil_timing_test.go
@@ -17,7 +17,7 @@ func BenchmarkToUnsafeString(b *testing.B) {
 				n += len(s)
 			}
 		}
-		atomic.AddUint64(&Sink, uint64(n))
+		Sink.Add(uint64(n))
 	})
 }
 
@@ -33,8 +33,8 @@ func BenchmarkToUnsafeBytes(b *testing.B) {
 				n += len(s)
 			}
 		}
-		atomic.AddUint64(&Sink, uint64(n))
+		Sink.Add(uint64(n))
 	})
 }
 
-var Sink uint64
+var Sink atomic.Uint64
diff --git a/lib/bytesutil/fast_string_matcher.go b/lib/bytesutil/fast_string_matcher.go
index 1b8d02b93..5d2b0643e 100644
--- a/lib/bytesutil/fast_string_matcher.go
+++ b/lib/bytesutil/fast_string_matcher.go
@@ -13,7 +13,7 @@ import (
 // It caches string match results and returns them back on the next calls
 // without calling the matchFunc, which may be expensive.
 type FastStringMatcher struct {
-	lastCleanupTime uint64
+	lastCleanupTime atomic.Uint64
 
 	m sync.Map
 
@@ -21,7 +21,7 @@ type FastStringMatcher struct {
 }
 
 type fsmEntry struct {
-	lastAccessTime uint64
+	lastAccessTime atomic.Uint64
 	ok             bool
 }
 
@@ -29,10 +29,11 @@ type fsmEntry struct {
 //
 // matchFunc must return the same result for the same input.
 func NewFastStringMatcher(matchFunc func(s string) bool) *FastStringMatcher {
-	return &FastStringMatcher{
-		lastCleanupTime: fasttime.UnixTimestamp(),
-		matchFunc:       matchFunc,
+	fsm := &FastStringMatcher{
+		matchFunc: matchFunc,
 	}
+	fsm.lastCleanupTime.Store(fasttime.UnixTimestamp())
+	return fsm
 }
 
 // Match applies matchFunc to s and returns the result.
@@ -46,19 +47,19 @@ func (fsm *FastStringMatcher) Match(s string) bool {
 	if ok {
 		// Fast path - s match result is found in the cache.
 		e := v.(*fsmEntry)
-		if atomic.LoadUint64(&e.lastAccessTime)+10 < ct {
+		if e.lastAccessTime.Load()+10 < ct {
 			// Reduce the frequency of e.lastAccessTime update to once per 10 seconds
 			// in order to improve the fast path speed on systems with many CPU cores.
-			atomic.StoreUint64(&e.lastAccessTime, ct)
+			e.lastAccessTime.Store(ct)
 		}
 		return e.ok
 	}
 	// Slow path - run matchFunc for s and store the result in the cache.
 	b := fsm.matchFunc(s)
 	e := &fsmEntry{
-		lastAccessTime: ct,
-		ok:             b,
+		ok: b,
 	}
+	e.lastAccessTime.Store(ct)
 	// Make a copy of s in order to limit memory usage to the s length,
 	// since the s may point to bigger string.
 	// This also protects from the case when s contains unsafe string, which points to a temporary byte slice.
@@ -72,7 +73,7 @@ func (fsm *FastStringMatcher) Match(s string) bool {
 		deadline := ct - uint64(cacheExpireDuration.Seconds())
 		m.Range(func(k, v interface{}) bool {
 			e := v.(*fsmEntry)
-			if atomic.LoadUint64(&e.lastAccessTime) < deadline {
+			if e.lastAccessTime.Load() < deadline {
 				m.Delete(k)
 			}
 			return true
@@ -82,13 +83,13 @@ func (fsm *FastStringMatcher) Match(s string) bool {
 	return b
 }
 
-func needCleanup(lastCleanupTime *uint64, currentTime uint64) bool {
-	lct := atomic.LoadUint64(lastCleanupTime)
+func needCleanup(lastCleanupTime *atomic.Uint64, currentTime uint64) bool {
+	lct := lastCleanupTime.Load()
 	if lct+61 >= currentTime {
 		return false
 	}
 	// Atomically compare and swap the current time with the lastCleanupTime
 	// in order to guarantee that only a single goroutine out of multiple
 	// concurrently executing goroutines gets true from the call.
-	return atomic.CompareAndSwapUint64(lastCleanupTime, lct, currentTime)
+	return lastCleanupTime.CompareAndSwap(lct, currentTime)
 }
diff --git a/lib/bytesutil/fast_string_matcher_test.go b/lib/bytesutil/fast_string_matcher_test.go
index f9c7e29f0..3643f7547 100644
--- a/lib/bytesutil/fast_string_matcher_test.go
+++ b/lib/bytesutil/fast_string_matcher_test.go
@@ -2,6 +2,7 @@ package bytesutil
 
 import (
 	"strings"
+	"sync/atomic"
 	"testing"
 )
 
@@ -27,18 +28,19 @@ func TestFastStringMatcher(t *testing.T) {
 func TestNeedCleanup(t *testing.T) {
 	f := func(lastCleanupTime, currentTime uint64, resultExpected bool) {
 		t.Helper()
-		lct := lastCleanupTime
+		var lct atomic.Uint64
+		lct.Store(lastCleanupTime)
 		result := needCleanup(&lct, currentTime)
 		if result != resultExpected {
 			t.Fatalf("unexpected result for needCleanup(%d, %d); got %v; want %v", lastCleanupTime, currentTime, result, resultExpected)
 		}
 		if result {
-			if lct != currentTime {
-				t.Fatalf("unexpected value for lct; got %d; want currentTime=%d", lct, currentTime)
+			if n := lct.Load(); n != currentTime {
+				t.Fatalf("unexpected value for lct; got %d; want currentTime=%d", n, currentTime)
 			}
 		} else {
-			if lct != lastCleanupTime {
-				t.Fatalf("unexpected value for lct; got %d; want lastCleanupTime=%d", lct, lastCleanupTime)
+			if n := lct.Load(); n != lastCleanupTime {
+				t.Fatalf("unexpected value for lct; got %d; want lastCleanupTime=%d", n, lastCleanupTime)
 			}
 		}
 	}
diff --git a/lib/bytesutil/fast_string_matcher_timing_test.go b/lib/bytesutil/fast_string_matcher_timing_test.go
index 7ccd51eed..49e6708c0 100644
--- a/lib/bytesutil/fast_string_matcher_timing_test.go
+++ b/lib/bytesutil/fast_string_matcher_timing_test.go
@@ -2,7 +2,6 @@ package bytesutil
 
 import (
 	"strings"
-	"sync/atomic"
 	"testing"
 )
 
@@ -28,6 +27,6 @@ func benchmarkFastStringMatcher(b *testing.B, s string) {
 				n++
 			}
 		}
-		atomic.AddUint64(&GlobalSink, n)
+		GlobalSink.Add(n)
 	})
 }
diff --git a/lib/bytesutil/fast_string_transformer.go b/lib/bytesutil/fast_string_transformer.go
index 4dcc930d1..342071cc9 100644
--- a/lib/bytesutil/fast_string_transformer.go
+++ b/lib/bytesutil/fast_string_transformer.go
@@ -13,7 +13,7 @@ import (
 // It caches transformed strings and returns them back on the next calls
 // without calling the transformFunc, which may be expensive.
 type FastStringTransformer struct {
-	lastCleanupTime uint64
+	lastCleanupTime atomic.Uint64
 
 	m sync.Map
 
@@ -21,7 +21,7 @@ type FastStringTransformer struct {
 }
 
 type fstEntry struct {
-	lastAccessTime uint64
+	lastAccessTime atomic.Uint64
 	s              string
 }
 
@@ -29,10 +29,11 @@ type fstEntry struct {
 //
 // transformFunc must return the same result for the same input.
 func NewFastStringTransformer(transformFunc func(s string) string) *FastStringTransformer {
-	return &FastStringTransformer{
-		lastCleanupTime: fasttime.UnixTimestamp(),
-		transformFunc:   transformFunc,
+	fst := &FastStringTransformer{
+		transformFunc: transformFunc,
 	}
+	fst.lastCleanupTime.Store(fasttime.UnixTimestamp())
+	return fst
 }
 
 // Transform applies transformFunc to s and returns the result.
@@ -52,10 +53,10 @@ func (fst *FastStringTransformer) Transform(s string) string {
 	if ok {
 		// Fast path - the transformed s is found in the cache.
 		e := v.(*fstEntry)
-		if atomic.LoadUint64(&e.lastAccessTime)+10 < ct {
+		if e.lastAccessTime.Load()+10 < ct {
 			// Reduce the frequency of e.lastAccessTime update to once per 10 seconds
 			// in order to improve the fast path speed on systems with many CPU cores.
-			atomic.StoreUint64(&e.lastAccessTime, ct)
+			e.lastAccessTime.Store(ct)
 		}
 		return e.s
 	}
@@ -72,9 +73,9 @@ func (fst *FastStringTransformer) Transform(s string) string {
 		sTransformed = s
 	}
 	e := &fstEntry{
-		lastAccessTime: ct,
-		s:              sTransformed,
+		s: sTransformed,
 	}
+	e.lastAccessTime.Store(ct)
 	fst.m.Store(s, e)
 
 	if needCleanup(&fst.lastCleanupTime, ct) {
@@ -83,7 +84,7 @@ func (fst *FastStringTransformer) Transform(s string) string {
 		deadline := ct - uint64(cacheExpireDuration.Seconds())
 		m.Range(func(k, v interface{}) bool {
 			e := v.(*fstEntry)
-			if atomic.LoadUint64(&e.lastAccessTime) < deadline {
+			if e.lastAccessTime.Load() < deadline {
 				m.Delete(k)
 			}
 			return true
diff --git a/lib/bytesutil/fast_string_transformer_timing_test.go b/lib/bytesutil/fast_string_transformer_timing_test.go
index b23fdbc05..793682934 100644
--- a/lib/bytesutil/fast_string_transformer_timing_test.go
+++ b/lib/bytesutil/fast_string_transformer_timing_test.go
@@ -24,8 +24,8 @@ func benchmarkFastStringTransformer(b *testing.B, s string) {
 			sTransformed := fst.Transform(s)
 			n += uint64(len(sTransformed))
 		}
-		atomic.AddUint64(&GlobalSink, n)
+		GlobalSink.Add(n)
 	})
 }
 
-var GlobalSink uint64
+var GlobalSink atomic.Uint64
diff --git a/lib/bytesutil/internstring.go b/lib/bytesutil/internstring.go
index a9521245e..e62445ec3 100644
--- a/lib/bytesutil/internstring.go
+++ b/lib/bytesutil/internstring.go
@@ -42,19 +42,19 @@ func InternString(s string) string {
 	ct := fasttime.UnixTimestamp()
 	if v, ok := internStringsMap.Load(s); ok {
 		e := v.(*ismEntry)
-		if atomic.LoadUint64(&e.lastAccessTime)+10 < ct {
+		if e.lastAccessTime.Load()+10 < ct {
 			// Reduce the frequency of e.lastAccessTime update to once per 10 seconds
 			// in order to improve the fast path speed on systems with many CPU cores.
-			atomic.StoreUint64(&e.lastAccessTime, ct)
+			e.lastAccessTime.Store(ct)
 		}
 		return e.s
 	}
 	// Make a new copy for s in order to remove references from possible bigger string s refers to.
 	sCopy := strings.Clone(s)
 	e := &ismEntry{
-		lastAccessTime: ct,
-		s:              sCopy,
+		s: sCopy,
 	}
+	e.lastAccessTime.Store(ct)
 	internStringsMap.Store(sCopy, e)
 
 	if needCleanup(&internStringsMapLastCleanupTime, ct) {
@@ -63,7 +63,7 @@ func InternString(s string) string {
 		deadline := ct - uint64(cacheExpireDuration.Seconds())
 		m.Range(func(k, v interface{}) bool {
 			e := v.(*ismEntry)
-			if atomic.LoadUint64(&e.lastAccessTime) < deadline {
+			if e.lastAccessTime.Load() < deadline {
 				m.Delete(k)
 			}
 			return true
@@ -74,11 +74,11 @@ func InternString(s string) string {
 }
 
 type ismEntry struct {
-	lastAccessTime uint64
+	lastAccessTime atomic.Uint64
 	s              string
 }
 
 var (
 	internStringsMap                sync.Map
-	internStringsMapLastCleanupTime uint64
+	internStringsMapLastCleanupTime atomic.Uint64
 )
diff --git a/lib/decimal/decimal_timing_test.go b/lib/decimal/decimal_timing_test.go
index 9dce15dee..061a18ef3 100644
--- a/lib/decimal/decimal_timing_test.go
+++ b/lib/decimal/decimal_timing_test.go
@@ -51,7 +51,7 @@ func benchmarkAppendDecimalToFloat(b *testing.B, a []int64, scale int16) {
 		var fa []float64
 		for pb.Next() {
 			fa = AppendDecimalToFloat(fa[:0], a, scale)
-			atomic.AddUint64(&Sink, uint64(len(fa)))
+			Sink.Add(uint64(len(fa)))
 		}
 	})
 }
@@ -101,7 +101,7 @@ func benchmarkAppendFloatToDecimal(b *testing.B, fa []float64) {
 			sink += uint64(len(da))
 			sink += uint64(e)
 		}
-		atomic.AddUint64(&Sink, sink)
+		Sink.Add(sink)
 	})
 }
 
@@ -138,10 +138,10 @@ func BenchmarkFromFloat(b *testing.B) {
 					sink += uint64(v)
 					sink += uint64(e)
 				}
-				atomic.AddUint64(&Sink, sink)
+				Sink.Add(sink)
 			})
 		})
 	}
 }
 
-var Sink uint64
+var Sink atomic.Uint64
diff --git a/lib/encoding/encoding_timing_test.go b/lib/encoding/encoding_timing_test.go
index fbfed7cf6..76843bd19 100644
--- a/lib/encoding/encoding_timing_test.go
+++ b/lib/encoding/encoding_timing_test.go
@@ -18,12 +18,12 @@ func BenchmarkMarshalGaugeArray(b *testing.B) {
 			if mt != MarshalTypeZSTDNearestDelta {
 				panic(fmt.Errorf("unexpected marshal type; got %d; expecting %d", mt, MarshalTypeZSTDNearestDelta))
 			}
-			atomic.AddUint64(&Sink, uint64(len(dst)))
+			Sink.Add(uint64(len(dst)))
 		}
 	})
 }
 
-var Sink uint64
+var Sink atomic.Uint64
 
 func BenchmarkUnmarshalGaugeArray(b *testing.B) {
 	b.ReportAllocs()
@@ -36,7 +36,7 @@ func BenchmarkUnmarshalGaugeArray(b *testing.B) {
 			if err != nil {
 				panic(fmt.Errorf("cannot unmarshal gauge array: %w", err))
 			}
-			atomic.AddUint64(&Sink, uint64(len(dst)))
+			Sink.Add(uint64(len(dst)))
 		}
 	})
 }
@@ -68,7 +68,7 @@ func BenchmarkMarshalDeltaConstArray(b *testing.B) {
 			if mt != MarshalTypeDeltaConst {
 				panic(fmt.Errorf("unexpected marshal type; got %d; expecting %d", mt, MarshalTypeDeltaConst))
 			}
-			atomic.AddUint64(&Sink, uint64(len(dst)))
+			Sink.Add(uint64(len(dst)))
 		}
 	})
 }
@@ -84,7 +84,7 @@ func BenchmarkUnmarshalDeltaConstArray(b *testing.B) {
 			if err != nil {
 				panic(fmt.Errorf("cannot unmarshal delta const array: %w", err))
 			}
-			atomic.AddUint64(&Sink, uint64(len(dst)))
+			Sink.Add(uint64(len(dst)))
 		}
 	})
 }
@@ -115,7 +115,7 @@ func BenchmarkMarshalConstArray(b *testing.B) {
 			if mt != MarshalTypeConst {
 				panic(fmt.Errorf("unexpected marshal type; got %d; expecting %d", mt, MarshalTypeConst))
 			}
-			atomic.AddUint64(&Sink, uint64(len(dst)))
+			Sink.Add(uint64(len(dst)))
 		}
 	})
 }
@@ -131,7 +131,7 @@ func BenchmarkUnmarshalConstArray(b *testing.B) {
 			if err != nil {
 				panic(fmt.Errorf("cannot unmarshal const array: %w", err))
 			}
-			atomic.AddUint64(&Sink, uint64(len(dst)))
+			Sink.Add(uint64(len(dst)))
 		}
 	})
 }
@@ -160,7 +160,7 @@ func BenchmarkMarshalZeroConstArray(b *testing.B) {
 			if mt != MarshalTypeConst {
 				panic(fmt.Errorf("unexpected marshal type; got %d; expecting %d", mt, MarshalTypeConst))
 			}
-			atomic.AddUint64(&Sink, uint64(len(dst)))
+			Sink.Add(uint64(len(dst)))
 		}
 	})
 }
@@ -176,7 +176,7 @@ func BenchmarkUnmarshalZeroConstArray(b *testing.B) {
 			if err != nil {
 				panic(fmt.Errorf("cannot unmarshal zero const array: %w", err))
 			}
-			atomic.AddUint64(&Sink, uint64(len(dst)))
+			Sink.Add(uint64(len(dst)))
 		}
 	})
 }
@@ -199,7 +199,7 @@ func BenchmarkMarshalInt64Array(b *testing.B) {
 			if mt != benchMarshalType {
 				panic(fmt.Errorf("unexpected marshal type; got %d; expecting %d", mt, benchMarshalType))
 			}
-			atomic.AddUint64(&Sink, uint64(len(dst)))
+			Sink.Add(uint64(len(dst)))
 		}
 	})
 }
@@ -215,7 +215,7 @@ func BenchmarkUnmarshalInt64Array(b *testing.B) {
 			if err != nil {
 				panic(fmt.Errorf("cannot unmarshal int64 array: %w", err))
 			}
-			atomic.AddUint64(&Sink, uint64(len(dst)))
+			Sink.Add(uint64(len(dst)))
 		}
 	})
 }
diff --git a/lib/encoding/int_timing_test.go b/lib/encoding/int_timing_test.go
index 0beb2ba49..1316ef514 100644
--- a/lib/encoding/int_timing_test.go
+++ b/lib/encoding/int_timing_test.go
@@ -2,7 +2,6 @@ package encoding
 
 import (
 	"fmt"
-	"sync/atomic"
 	"testing"
 )
 
@@ -16,7 +15,7 @@ func BenchmarkMarshalUint64(b *testing.B) {
 			dst = MarshalUint64(dst[:0], sink)
 			sink += uint64(len(dst))
 		}
-		atomic.AddUint64(&Sink, sink)
+		Sink.Add(sink)
 	})
 }
 
@@ -29,7 +28,7 @@ func BenchmarkUnmarshalUint64(b *testing.B) {
 			v := UnmarshalUint64(testMarshaledUint64Data)
 			sink += v
 		}
-		atomic.AddUint64(&Sink, sink)
+		Sink.Add(sink)
 	})
 }
 
@@ -43,7 +42,7 @@ func BenchmarkMarshalInt64(b *testing.B) {
 			dst = MarshalInt64(dst[:0], int64(sink))
 			sink += uint64(len(dst))
 		}
-		atomic.AddUint64(&Sink, sink)
+		Sink.Add(sink)
 	})
 }
 
@@ -56,7 +55,7 @@ func BenchmarkUnmarshalInt64(b *testing.B) {
 			v := UnmarshalInt64(testMarshaledInt64Data)
 			sink += uint64(v)
 		}
-		atomic.AddUint64(&Sink, sink)
+		Sink.Add(sink)
 	})
 }
 
@@ -96,7 +95,7 @@ func benchmarkMarshalVarUint64s(b *testing.B, maxValue uint64) {
 			dst = MarshalVarUint64s(dst[:0], data)
 			sink += uint64(len(dst))
 		}
-		atomic.AddUint64(&Sink, sink)
+		Sink.Add(sink)
 	})
 }
 
@@ -136,7 +135,7 @@ func benchmarkMarshalVarInt64s(b *testing.B, maxValue int64) {
 			dst = MarshalVarInt64s(dst[:0], data)
 			sink += uint64(len(dst))
 		}
-		atomic.AddUint64(&Sink, sink)
+		Sink.Add(sink)
 	})
 }
 
@@ -182,7 +181,7 @@ func benchmarkUnmarshalVarUint64s(b *testing.B, maxValue uint64) {
 			}
 			sink += uint64(len(dst))
 		}
-		atomic.AddUint64(&Sink, sink)
+		Sink.Add(sink)
 	})
 }
 
@@ -228,7 +227,7 @@ func benchmarkUnmarshalVarInt64s(b *testing.B, maxValue int64) {
 			}
 			sink += uint64(len(dst))
 		}
-		atomic.AddUint64(&Sink, sink)
+		Sink.Add(sink)
 	})
 }
 
diff --git a/lib/encoding/nearest_delta2_timing_test.go b/lib/encoding/nearest_delta2_timing_test.go
index ee58e4c83..b10196168 100644
--- a/lib/encoding/nearest_delta2_timing_test.go
+++ b/lib/encoding/nearest_delta2_timing_test.go
@@ -2,7 +2,6 @@ package encoding
 
 import (
 	"fmt"
-	"sync/atomic"
 	"testing"
 )
 
@@ -21,7 +20,7 @@ func benchmarkMarshalInt64NearestDelta2(b *testing.B, precisionBits uint8) {
 		var dst []byte
 		for pb.Next() {
 			dst, _ = marshalInt64NearestDelta2(dst[:0], benchInt64Array, precisionBits)
-			atomic.AddUint64(&Sink, uint64(len(dst)))
+			Sink.Add(uint64(len(dst)))
 		}
 	})
 }
@@ -37,7 +36,7 @@ func BenchmarkUnmarshalInt64NearestDelta2(b *testing.B) {
 			if err != nil {
 				panic(fmt.Errorf("unexpected error: %w", err))
 			}
-			atomic.AddUint64(&Sink, uint64(len(dst)))
+			Sink.Add(uint64(len(dst)))
 		}
 	})
 }
diff --git a/lib/encoding/nearest_delta_timing_test.go b/lib/encoding/nearest_delta_timing_test.go
index 3eb0c3126..383e7e147 100644
--- a/lib/encoding/nearest_delta_timing_test.go
+++ b/lib/encoding/nearest_delta_timing_test.go
@@ -2,7 +2,6 @@ package encoding
 
 import (
 	"fmt"
-	"sync/atomic"
 	"testing"
 )
 
@@ -21,7 +20,7 @@ func benchmarkMarshalInt64NearestDelta(b *testing.B, precisionBits uint8) {
 		var dst []byte
 		for pb.Next() {
 			dst, _ = marshalInt64NearestDelta(dst[:0], benchInt64Array, precisionBits)
-			atomic.AddUint64(&Sink, uint64(len(dst)))
+			Sink.Add(uint64(len(dst)))
 		}
 	})
 }
@@ -37,7 +36,7 @@ func BenchmarkUnmarshalInt64NearestDelta(b *testing.B) {
 			if err != nil {
 				panic(fmt.Errorf("unexpected error: %w", err))
 			}
-			atomic.AddUint64(&Sink, uint64(len(dst)))
+			Sink.Add(uint64(len(dst)))
 		}
 	})
 }
diff --git a/lib/fasttime/fasttime.go b/lib/fasttime/fasttime.go
index f50a27f7f..cf44a51de 100644
--- a/lib/fasttime/fasttime.go
+++ b/lib/fasttime/fasttime.go
@@ -11,18 +11,22 @@ func init() {
 		defer ticker.Stop()
 		for tm := range ticker.C {
 			t := uint64(tm.Unix())
-			atomic.StoreUint64(&currentTimestamp, t)
+			currentTimestamp.Store(t)
 		}
 	}()
 }
 
-var currentTimestamp = uint64(time.Now().Unix())
+var currentTimestamp = func() *atomic.Uint64 {
+	var x atomic.Uint64
+	x.Store(uint64(time.Now().Unix()))
+	return &x
+}()
 
 // UnixTimestamp returns the current unix timestamp in seconds.
 //
 // It is faster than time.Now().Unix()
 func UnixTimestamp() uint64 {
-	return atomic.LoadUint64(&currentTimestamp)
+	return currentTimestamp.Load()
 }
 
 // UnixDate returns date from the current unix timestamp.
diff --git a/lib/fasttime/fasttime_timing_test.go b/lib/fasttime/fasttime_timing_test.go
index bd27085c8..becda781b 100644
--- a/lib/fasttime/fasttime_timing_test.go
+++ b/lib/fasttime/fasttime_timing_test.go
@@ -13,7 +13,7 @@ func BenchmarkUnixTimestamp(b *testing.B) {
 		for pb.Next() {
 			ts += UnixTimestamp()
 		}
-		atomic.StoreUint64(&Sink, ts)
+		Sink.Store(ts)
 	})
 }
 
@@ -24,9 +24,9 @@ func BenchmarkTimeNowUnix(b *testing.B) {
 		for pb.Next() {
 			ts += uint64(time.Now().Unix())
 		}
-		atomic.StoreUint64(&Sink, ts)
+		Sink.Store(ts)
 	})
 }
 
 // Sink should prevent from code elimination by optimizing compiler
-var Sink uint64
+var Sink atomic.Uint64
diff --git a/lib/flagutil/password.go b/lib/flagutil/password.go
index 8aff738f9..0830e8bbf 100644
--- a/lib/flagutil/password.go
+++ b/lib/flagutil/password.go
@@ -35,7 +35,7 @@ func NewPassword(name, description string) *Password {
 // If the flag value is file:///path/to/file or http://host/path ,
 // then its contents is automatically re-read from the given file or url
 type Password struct {
-	nextRefreshTimestamp uint64
+	nextRefreshTimestamp atomic.Uint64
 
 	value atomic.Pointer[string]
 
@@ -62,14 +62,14 @@ func (p *Password) maybeRereadPassword() {
 		return
 	}
 	tsCurr := fasttime.UnixTimestamp()
-	tsNext := atomic.LoadUint64(&p.nextRefreshTimestamp)
+	tsNext := p.nextRefreshTimestamp.Load()
 	if tsCurr < tsNext {
 		// Fast path - nothing to re-read
 		return
 	}
 
 	// Re-read password from p.sourcePath
-	atomic.StoreUint64(&p.nextRefreshTimestamp, tsCurr+2)
+	p.nextRefreshTimestamp.Store(tsCurr + 2)
 	s, err := fscore.ReadPasswordFromFileOrHTTP(p.sourcePath)
 	if err != nil {
 		// cannot use lib/logger, since it can be uninitialized yet
@@ -86,7 +86,7 @@ func (p *Password) String() string {
 
 // Set implements flag.Value interface.
 func (p *Password) Set(value string) error {
-	atomic.StoreUint64(&p.nextRefreshTimestamp, 0)
+	p.nextRefreshTimestamp.Store(0)
 	switch {
 	case strings.HasPrefix(value, "file://"):
 		p.sourcePath = strings.TrimPrefix(value, "file://")
diff --git a/lib/fs/fs.go b/lib/fs/fs.go
index a3521b051..980066b10 100644
--- a/lib/fs/fs.go
+++ b/lib/fs/fs.go
@@ -16,7 +16,7 @@ import (
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/logger"
 )
 
-var tmpFileNum uint64
+var tmpFileNum atomic.Uint64
 
 // MustSyncPath syncs contents of the given path.
 func MustSyncPath(path string) {
@@ -62,7 +62,7 @@ func MustWriteAtomic(path string, data []byte, canOverwrite bool) {
 	}
 
 	// Write data to a temporary file.
-	n := atomic.AddUint64(&tmpFileNum, 1)
+	n := tmpFileNum.Add(1)
 	tmpPath := fmt.Sprintf("%s.tmp.%d", path, n)
 	MustWriteSync(tmpPath, data)
 
@@ -207,7 +207,11 @@ func MustRemoveDirAtomic(dir string) {
 	MustSyncPath(parentDir)
 }
 
-var atomicDirRemoveCounter = uint64(time.Now().UnixNano())
+var atomicDirRemoveCounter = func() *atomic.Uint64 {
+	var x atomic.Uint64
+	x.Store(uint64(time.Now().UnixNano()))
+	return &x
+}()
 
 // MustReadDir reads directory entries at the given dir.
 func MustReadDir(dir string) []os.DirEntry {
diff --git a/lib/fs/fs_nix.go b/lib/fs/fs_nix.go
index fed424edb..634727f04 100644
--- a/lib/fs/fs_nix.go
+++ b/lib/fs/fs_nix.go
@@ -5,7 +5,6 @@ package fs
 import (
 	"fmt"
 	"os"
-	"sync/atomic"
 
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/logger"
 	"golang.org/x/sys/unix"
@@ -16,7 +15,7 @@ func freeSpace(stat unix.Statfs_t) uint64 {
 }
 
 func mustRemoveDirAtomic(dir string) {
-	n := atomic.AddUint64(&atomicDirRemoveCounter, 1)
+	n := atomicDirRemoveCounter.Add(1)
 	tmpDir := fmt.Sprintf("%s.must-remove.%d", dir, n)
 	if err := os.Rename(dir, tmpDir); err != nil {
 		logger.Panicf("FATAL: cannot move %s to %s: %s", dir, tmpDir, err)
diff --git a/lib/fs/fs_openbsd.go b/lib/fs/fs_openbsd.go
index 6f3d4904e..2f4574ea6 100644
--- a/lib/fs/fs_openbsd.go
+++ b/lib/fs/fs_openbsd.go
@@ -3,7 +3,6 @@ package fs
 import (
 	"fmt"
 	"os"
-	"sync/atomic"
 
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/logger"
 	"golang.org/x/sys/unix"
@@ -14,7 +13,7 @@ func freeSpace(stat unix.Statfs_t) uint64 {
 }
 
 func mustRemoveDirAtomic(dir string) {
-	n := atomic.AddUint64(&atomicDirRemoveCounter, 1)
+	n := atomicDirRemoveCounter.Add(1)
 	tmpDir := fmt.Sprintf("%s.must-remove.%d", dir, n)
 	if err := os.Rename(dir, tmpDir); err != nil {
 		logger.Panicf("FATAL: cannot move %s to %s: %s", dir, tmpDir, err)
diff --git a/lib/fs/fs_solaris.go b/lib/fs/fs_solaris.go
index 923f69e65..b3d880446 100644
--- a/lib/fs/fs_solaris.go
+++ b/lib/fs/fs_solaris.go
@@ -3,14 +3,13 @@ package fs
 import (
 	"fmt"
 	"os"
-	"sync/atomic"
 
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/logger"
 	"golang.org/x/sys/unix"
 )
 
 func mustRemoveDirAtomic(dir string) {
-	n := atomic.AddUint64(&atomicDirRemoveCounter, 1)
+	n := atomicDirRemoveCounter.Add(1)
 	tmpDir := fmt.Sprintf("%s.must-remove.%d", dir, n)
 	if err := os.Rename(dir, tmpDir); err != nil {
 		logger.Panicf("FATAL: cannot move %s to %s: %s", dir, tmpDir, err)
diff --git a/lib/fs/fs_windows.go b/lib/fs/fs_windows.go
index f455c78cf..af2076afd 100644
--- a/lib/fs/fs_windows.go
+++ b/lib/fs/fs_windows.go
@@ -5,7 +5,6 @@ import (
 	"os"
 	"reflect"
 	"sync"
-	"sync/atomic"
 	"unsafe"
 
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/logger"
@@ -25,7 +24,7 @@ func mustSyncPath(path string) {
 }
 
 func mustRemoveDirAtomic(dir string) {
-	n := atomic.AddUint64(&atomicDirRemoveCounter, 1)
+	n := atomicDirRemoveCounter.Add(1)
 	tmpDir := fmt.Sprintf("%s.must-remove.%d", dir, n)
 	if err := os.Rename(dir, tmpDir); err != nil {
 		logger.Panicf("FATAL: cannot move %s to %s: %s", dir, tmpDir, err)
diff --git a/lib/fs/reader_at.go b/lib/fs/reader_at.go
index 379c05a9e..061edae8b 100644
--- a/lib/fs/reader_at.go
+++ b/lib/fs/reader_at.go
@@ -32,8 +32,8 @@ type MustReadAtCloser interface {
 
 // ReaderAt implements rand-access reader.
 type ReaderAt struct {
-	readCalls uint64
-	readBytes uint64
+	readCalls atomic.Int64
+	readBytes atomic.Int64
 
 	// path contains the path to the file for reading
 	path string
@@ -81,8 +81,8 @@ func (r *ReaderAt) MustReadAt(p []byte, off int64) {
 		copy(p, src)
 	}
 	if r.useLocalStats {
-		atomic.AddUint64(&r.readCalls, 1)
-		atomic.AddUint64(&r.readBytes, uint64(len(p)))
+		r.readCalls.Add(1)
+		r.readBytes.Add(int64(len(p)))
 	} else {
 		readCalls.Inc()
 		readBytes.Add(len(p))
@@ -119,10 +119,10 @@ func (r *ReaderAt) MustClose() {
 	}
 
 	if r.useLocalStats {
-		readCalls.Add(int(r.readCalls))
-		readBytes.Add(int(r.readBytes))
-		r.readCalls = 0
-		r.readBytes = 0
+		readCalls.AddInt64(r.readCalls.Load())
+		readBytes.AddInt64(r.readBytes.Load())
+		r.readCalls.Store(0)
+		r.readBytes.Store(0)
 		r.useLocalStats = false
 	}
 }
diff --git a/lib/httpserver/httpserver.go b/lib/httpserver/httpserver.go
index 1f9f9fb1b..45ea74903 100644
--- a/lib/httpserver/httpserver.go
+++ b/lib/httpserver/httpserver.go
@@ -68,7 +68,7 @@ var (
 )
 
 type server struct {
-	shutdownDelayDeadline int64
+	shutdownDelayDeadline atomic.Int64
 	s                     *http.Server
 }
 
@@ -225,7 +225,7 @@ func stop(addr string) error {
 	}
 
 	deadline := time.Now().Add(*shutdownDelay).UnixNano()
-	atomic.StoreInt64(&s.shutdownDelayDeadline, deadline)
+	s.shutdownDelayDeadline.Store(deadline)
 	if *shutdownDelay > 0 {
 		// Sleep for a while until load balancer in front of the server
 		// notifies that "/health" endpoint returns non-OK responses.
@@ -339,7 +339,7 @@ func handlerWrapper(s *server, w http.ResponseWriter, r *http.Request, rh Reques
 	switch r.URL.Path {
 	case "/health":
 		h.Set("Content-Type", "text/plain; charset=utf-8")
-		deadline := atomic.LoadInt64(&s.shutdownDelayDeadline)
+		deadline := s.shutdownDelayDeadline.Load()
 		if deadline <= 0 {
 			w.Write([]byte("OK"))
 			return
diff --git a/lib/lrucache/lrucache.go b/lib/lrucache/lrucache.go
index 3d545cb4a..e3199e95e 100644
--- a/lib/lrucache/lrucache.go
+++ b/lib/lrucache/lrucache.go
@@ -149,14 +149,11 @@ func (c *Cache) cleanByTimeout() {
 }
 
 type cache struct {
-	// Atomically updated fields must go first in the struct, so they are properly
-	// aligned to 8 bytes on 32-bit architectures.
-	// See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/212
-	requests uint64
-	misses   uint64
+	requests atomic.Uint64
+	misses   atomic.Uint64
 
 	// sizeBytes contains an approximate size for all the blocks stored in the cache.
-	sizeBytes int64
+	sizeBytes atomic.Int64
 
 	// getMaxSizeBytes() is a callback, which returns the maximum allowed cache size in bytes.
 	getMaxSizeBytes func() int
@@ -204,7 +201,7 @@ func newCache(getMaxSizeBytes func() int) *cache {
 }
 
 func (c *cache) updateSizeBytes(n int) {
-	atomic.AddInt64(&c.sizeBytes, int64(n))
+	c.sizeBytes.Add(int64(n))
 }
 
 func (c *cache) cleanByTimeout() {
@@ -223,13 +220,13 @@ func (c *cache) cleanByTimeout() {
 }
 
 func (c *cache) GetEntry(k string) Entry {
-	atomic.AddUint64(&c.requests, 1)
+	c.requests.Add(1)
 	c.mu.Lock()
 	defer c.mu.Unlock()
 
 	ce := c.m[k]
 	if ce == nil {
-		atomic.AddUint64(&c.misses, 1)
+		c.misses.Add(1)
 		return nil
 	}
 	currentTime := fasttime.UnixTimestamp()
@@ -277,7 +274,7 @@ func (c *cache) Len() int {
 }
 
 func (c *cache) SizeBytes() int {
-	return int(atomic.LoadInt64(&c.sizeBytes))
+	return int(c.sizeBytes.Load())
 }
 
 func (c *cache) SizeMaxBytes() int {
@@ -285,11 +282,11 @@ func (c *cache) SizeMaxBytes() int {
 }
 
 func (c *cache) Requests() uint64 {
-	return atomic.LoadUint64(&c.requests)
+	return c.requests.Load()
 }
 
 func (c *cache) Misses() uint64 {
-	return atomic.LoadUint64(&c.misses)
+	return c.misses.Load()
 }
 
 // lastAccessHeap implements heap.Interface
diff --git a/lib/netutil/conn.go b/lib/netutil/conn.go
index 651752f74..41aefb0b9 100644
--- a/lib/netutil/conn.go
+++ b/lib/netutil/conn.go
@@ -44,10 +44,7 @@ func (cm *connMetrics) init(ms *metrics.Set, group, name, addr string) {
 }
 
 type statConn struct {
-	// Move atomic counters to the top of struct in order to properly align them on 32-bit arch.
-	// See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/212
-
-	closeCalls uint64
+	closeCalls atomic.Uint64
 
 	net.Conn
 
@@ -90,7 +87,7 @@ func (sc *statConn) Write(p []byte) (int, error) {
 }
 
 func (sc *statConn) Close() error {
-	n := atomic.AddUint64(&sc.closeCalls, 1)
+	n := sc.closeCalls.Add(1)
 	if n > 1 {
 		// The connection has been already closed.
 		return nil
diff --git a/lib/promscrape/discovery/kubernetes/api_watcher.go b/lib/promscrape/discovery/kubernetes/api_watcher.go
index 48544f264..f1c4470b2 100644
--- a/lib/promscrape/discovery/kubernetes/api_watcher.go
+++ b/lib/promscrape/discovery/kubernetes/api_watcher.go
@@ -104,9 +104,9 @@ func newAPIWatcher(apiServer string, ac *promauth.Config, sdc *SDConfig, swcFunc
 }
 
 func (aw *apiWatcher) mustStart() {
-	atomic.AddInt32(&aw.gw.apiWatcherInflightStartCalls, 1)
+	aw.gw.apiWatcherInflightStartCalls.Add(1)
 	aw.gw.startWatchersForRole(aw.role, aw)
-	atomic.AddInt32(&aw.gw.apiWatcherInflightStartCalls, -1)
+	aw.gw.apiWatcherInflightStartCalls.Add(-1)
 }
 
 func (aw *apiWatcher) updateSwosCount(multiplier int, swosByKey map[string][]interface{}) {
@@ -214,15 +214,15 @@ func (aw *apiWatcher) getScrapeWorkObjects() []interface{} {
 type groupWatcher struct {
 	// The number of in-flight apiWatcher.mustStart() calls for the given groupWatcher.
 	// This field is used by groupWatchersCleaner() in order to determine when the given groupWatcher can be stopped.
-	apiWatcherInflightStartCalls int32
+	apiWatcherInflightStartCalls atomic.Int32
 
 	// Old Kubernetes doesn't support /apis/networking.k8s.io/v1/, so /apis/networking.k8s.io/v1beta1/ must be used instead.
 	// This flag is used for automatic substitution of v1 API path with v1beta1 API path during requests to apiServer.
-	useNetworkingV1Beta1 uint32
+	useNetworkingV1Beta1 atomic.Bool
 
 	// Old Kubernetes doesn't support /apis/discovery.k8s.io/v1/, so discovery.k8s.io/v1beta1/ must be used instead.
 	// This flag is used for automatic substitution of v1 API path with v1beta1 API path during requests to apiServer.
-	useDiscoveryV1Beta1 uint32
+	useDiscoveryV1Beta1 atomic.Bool
 
 	apiServer          string
 	namespaces         []string
@@ -343,7 +343,7 @@ func groupWatchersCleaner() {
 				awsTotal += len(uw.aws) + len(uw.awsPending)
 			}
 
-			if awsTotal == 0 && atomic.LoadInt32(&gw.apiWatcherInflightStartCalls) == 0 {
+			if awsTotal == 0 && gw.apiWatcherInflightStartCalls.Load() == 0 {
 				// There are no API watchers subscribed to gw and there are no in-flight apiWatcher.mustStart() calls.
 				// Stop all the urlWatcher instances at gw and drop gw from groupWatchers in this case,
 				// but do it only on the second iteration in order to reduce urlWatcher churn
@@ -471,11 +471,11 @@ func (gw *groupWatcher) startWatchersForRole(role string, aw *apiWatcher) {
 
 // doRequest performs http request to the given requestURL.
 func (gw *groupWatcher) doRequest(ctx context.Context, requestURL string) (*http.Response, error) {
-	if strings.Contains(requestURL, "/apis/networking.k8s.io/v1/") && atomic.LoadUint32(&gw.useNetworkingV1Beta1) == 1 {
+	if strings.Contains(requestURL, "/apis/networking.k8s.io/v1/") && gw.useNetworkingV1Beta1.Load() {
 		// Update networking URL for old Kubernetes API, which supports only v1beta1 path.
 		requestURL = strings.Replace(requestURL, "/apis/networking.k8s.io/v1/", "/apis/networking.k8s.io/v1beta1/", 1)
 	}
-	if strings.Contains(requestURL, "/apis/discovery.k8s.io/v1/") && atomic.LoadUint32(&gw.useDiscoveryV1Beta1) == 1 {
+	if strings.Contains(requestURL, "/apis/discovery.k8s.io/v1/") && gw.useDiscoveryV1Beta1.Load() {
 		// Update discovery URL for old Kubernetes API, which supports only v1beta1 path.
 		requestURL = strings.Replace(requestURL, "/apis/discovery.k8s.io/v1/", "/apis/discovery.k8s.io/v1beta1/", 1)
 	}
@@ -491,12 +491,12 @@ func (gw *groupWatcher) doRequest(ctx context.Context, requestURL string) (*http
 		return nil, err
 	}
 	if resp.StatusCode == http.StatusNotFound {
-		if strings.Contains(requestURL, "/apis/networking.k8s.io/v1/") && atomic.LoadUint32(&gw.useNetworkingV1Beta1) == 0 {
-			atomic.StoreUint32(&gw.useNetworkingV1Beta1, 1)
+		if strings.Contains(requestURL, "/apis/networking.k8s.io/v1/") && !gw.useNetworkingV1Beta1.Load() {
+			gw.useNetworkingV1Beta1.Store(true)
 			return gw.doRequest(ctx, requestURL)
 		}
-		if strings.Contains(requestURL, "/apis/discovery.k8s.io/v1/") && atomic.LoadUint32(&gw.useDiscoveryV1Beta1) == 0 {
-			atomic.StoreUint32(&gw.useDiscoveryV1Beta1, 1)
+		if strings.Contains(requestURL, "/apis/discovery.k8s.io/v1/") && !gw.useDiscoveryV1Beta1.Load() {
+			gw.useDiscoveryV1Beta1.Store(true)
 			return gw.doRequest(ctx, requestURL)
 		}
 	}
diff --git a/lib/promscrape/scraper.go b/lib/promscrape/scraper.go
index 183dcd0e1..87b52a5b3 100644
--- a/lib/promscrape/scraper.go
+++ b/lib/promscrape/scraper.go
@@ -79,9 +79,9 @@ func Stop() {
 var (
 	globalStopChan chan struct{}
 	scraperWG      sync.WaitGroup
-	// PendingScrapeConfigs - zero value means, that
-	// all scrapeConfigs are inited and ready for work.
-	PendingScrapeConfigs int32
+
+	// PendingScrapeConfigs - zero value means, that all scrapeConfigs are inited and ready for work.
+	PendingScrapeConfigs atomic.Int32
 
 	// configData contains -promscrape.config data
 	configData atomic.Pointer[[]byte]
@@ -225,7 +225,7 @@ func newScrapeConfigs(pushData func(at *auth.Token, wr *prompbmarshal.WriteReque
 }
 
 func (scs *scrapeConfigs) add(name string, checkInterval time.Duration, getScrapeWork func(cfg *Config, swsPrev []*ScrapeWork) []*ScrapeWork) {
-	atomic.AddInt32(&PendingScrapeConfigs, 1)
+	PendingScrapeConfigs.Add(1)
 	scfg := &scrapeConfig{
 		name:          name,
 		pushData:      scs.pushData,
@@ -292,7 +292,7 @@ func (scfg *scrapeConfig) run(globalStopCh <-chan struct{}) {
 		}
 	}
 	updateScrapeWork(cfg)
-	atomic.AddInt32(&PendingScrapeConfigs, -1)
+	PendingScrapeConfigs.Add(-1)
 
 	for {
 
diff --git a/lib/promscrape/statconn.go b/lib/promscrape/statconn.go
index 6167646da..296f8d1d6 100644
--- a/lib/promscrape/statconn.go
+++ b/lib/promscrape/statconn.go
@@ -56,7 +56,7 @@ var (
 )
 
 type statConn struct {
-	closed uint64
+	closed atomic.Int32
 	net.Conn
 }
 
@@ -82,7 +82,7 @@ func (sc *statConn) Write(p []byte) (int, error) {
 
 func (sc *statConn) Close() error {
 	err := sc.Conn.Close()
-	if atomic.AddUint64(&sc.closed, 1) == 1 {
+	if sc.closed.Add(1) == 1 {
 		conns.Dec()
 	}
 	return err
diff --git a/lib/snapshot/snapshotutil/snapshotutil.go b/lib/snapshot/snapshotutil/snapshotutil.go
index 9c55fe4ac..77dcf910f 100644
--- a/lib/snapshot/snapshotutil/snapshotutil.go
+++ b/lib/snapshot/snapshotutil/snapshotutil.go
@@ -41,7 +41,11 @@ func NewName() string {
 }
 
 func nextSnapshotIdx() uint64 {
-	return atomic.AddUint64(&snapshotIdx, 1)
+	return snapshotIdx.Add(1)
 }
 
-var snapshotIdx = uint64(time.Now().UnixNano())
+var snapshotIdx = func() *atomic.Uint64 {
+	var x atomic.Uint64
+	x.Store(uint64(time.Now().UnixNano()))
+	return &x
+}()
diff --git a/lib/workingsetcache/cache.go b/lib/workingsetcache/cache.go
index cc27fcb08..16dffe310 100644
--- a/lib/workingsetcache/cache.go
+++ b/lib/workingsetcache/cache.go
@@ -43,7 +43,7 @@ type Cache struct {
 	// In this case using prev would result in RAM waste,
 	// it is better to use only curr cache with doubled size.
 	// After the process of switching, this flag will be set to whole.
-	mode uint32
+	mode atomic.Uint32
 
 	// The maxBytes value passed to New() or to Load().
 	maxBytes int
@@ -110,7 +110,7 @@ func newCacheInternal(curr, prev *fastcache.Cache, mode, maxBytes int) *Cache {
 	c.curr.Store(curr)
 	c.prev.Store(prev)
 	c.stopCh = make(chan struct{})
-	c.setMode(mode)
+	c.mode.Store(uint32(mode))
 	return &c
 }
 
@@ -143,7 +143,7 @@ func (c *Cache) expirationWatcher(expireDuration time.Duration) {
 		case <-t.C:
 		}
 		c.mu.Lock()
-		if atomic.LoadUint32(&c.mode) != split {
+		if c.mode.Load() != split {
 			// Stop the expirationWatcher on non-split mode.
 			c.mu.Unlock()
 			return
@@ -183,7 +183,7 @@ func (c *Cache) prevCacheWatcher() {
 		case <-t.C:
 		}
 		c.mu.Lock()
-		if atomic.LoadUint32(&c.mode) != split {
+		if c.mode.Load() != split {
 			// Do nothing in non-split mode.
 			c.mu.Unlock()
 			return
@@ -227,7 +227,7 @@ func (c *Cache) cacheSizeWatcher() {
 			return
 		case <-t.C:
 		}
-		if c.loadMode() != split {
+		if c.mode.Load() != split {
 			continue
 		}
 		var cs fastcache.Stats
@@ -252,7 +252,7 @@ func (c *Cache) cacheSizeWatcher() {
 	// 6) drop prev cache
 
 	c.mu.Lock()
-	c.setMode(switching)
+	c.mode.Store(switching)
 	prev := c.prev.Load()
 	curr := c.curr.Load()
 	c.prev.Store(curr)
@@ -280,7 +280,7 @@ func (c *Cache) cacheSizeWatcher() {
 	}
 
 	c.mu.Lock()
-	c.setMode(whole)
+	c.mode.Store(whole)
 	prev = c.prev.Load()
 	c.prev.Store(fastcache.New(1024))
 	cs.Reset()
@@ -318,15 +318,7 @@ func (c *Cache) Reset() {
 	updateCacheStatsHistory(&c.csHistory, &cs)
 	curr.Reset()
 	// Reset the mode to `split` in the hope the working set size becomes smaller after the reset.
-	c.setMode(split)
-}
-
-func (c *Cache) setMode(mode int) {
-	atomic.StoreUint32(&c.mode, uint32(mode))
-}
-
-func (c *Cache) loadMode() int {
-	return int(atomic.LoadUint32(&c.mode))
+	c.mode.Store(split)
 }
 
 // UpdateStats updates fcs with cache stats.
@@ -374,7 +366,7 @@ func (c *Cache) Get(dst, key []byte) []byte {
 		// Fast path - the entry is found in the current cache.
 		return result
 	}
-	if c.loadMode() == whole {
+	if c.mode.Load() == whole {
 		// Nothing found.
 		return result
 	}
@@ -397,7 +389,7 @@ func (c *Cache) Has(key []byte) bool {
 	if curr.Has(key) {
 		return true
 	}
-	if c.loadMode() == whole {
+	if c.mode.Load() == whole {
 		return false
 	}
 	prev := c.prev.Load()
@@ -428,7 +420,7 @@ func (c *Cache) GetBig(dst, key []byte) []byte {
 		// Fast path - the entry is found in the current cache.
 		return result
 	}
-	if c.loadMode() == whole {
+	if c.mode.Load() == whole {
 		// Nothing found.
 		return result
 	}
diff --git a/vendor/github.com/VictoriaMetrics/metrics/counter.go b/vendor/github.com/VictoriaMetrics/metrics/counter.go
index 25973a8a5..1076e80c2 100644
--- a/vendor/github.com/VictoriaMetrics/metrics/counter.go
+++ b/vendor/github.com/VictoriaMetrics/metrics/counter.go
@@ -42,6 +42,11 @@ func (c *Counter) Add(n int) {
 	atomic.AddUint64(&c.n, uint64(n))
 }
 
+// AddInt64 adds n to c.
+func (c *Counter) AddInt64(n int64) {
+	atomic.AddUint64(&c.n, uint64(n))
+}
+
 // Get returns the current value for c.
 func (c *Counter) Get() uint64 {
 	return atomic.LoadUint64(&c.n)
diff --git a/vendor/modules.txt b/vendor/modules.txt
index d183be4bf..878000848 100644
--- a/vendor/modules.txt
+++ b/vendor/modules.txt
@@ -99,7 +99,7 @@ github.com/VictoriaMetrics/easyproto
 # github.com/VictoriaMetrics/fastcache v1.12.2
 ## explicit; go 1.13
 github.com/VictoriaMetrics/fastcache
-# github.com/VictoriaMetrics/metrics v1.32.0
+# github.com/VictoriaMetrics/metrics v1.33.0
 ## explicit; go 1.17
 github.com/VictoriaMetrics/metrics
 # github.com/VictoriaMetrics/metricsql v0.74.0